Upload folder using huggingface_hub
Browse files- checkpoint-361/README.md +202 -0
- checkpoint-361/adapter_config.json +40 -0
- checkpoint-361/adapter_model.safetensors +3 -0
- checkpoint-361/optimizer.pt +3 -0
- checkpoint-361/rng_state.pth +3 -0
- checkpoint-361/scheduler.pt +3 -0
- checkpoint-361/trainer_state.json +2560 -0
- checkpoint-361/training_args.bin +3 -0
checkpoint-361/README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: rinna/llama-3-youko-8b
|
| 3 |
+
library_name: peft
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Card for Model ID
|
| 7 |
+
|
| 8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Details
|
| 13 |
+
|
| 14 |
+
### Model Description
|
| 15 |
+
|
| 16 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- **Developed by:** [More Information Needed]
|
| 21 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 22 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 23 |
+
- **Model type:** [More Information Needed]
|
| 24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 25 |
+
- **License:** [More Information Needed]
|
| 26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 27 |
+
|
| 28 |
+
### Model Sources [optional]
|
| 29 |
+
|
| 30 |
+
<!-- Provide the basic links for the model. -->
|
| 31 |
+
|
| 32 |
+
- **Repository:** [More Information Needed]
|
| 33 |
+
- **Paper [optional]:** [More Information Needed]
|
| 34 |
+
- **Demo [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
## Uses
|
| 37 |
+
|
| 38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 39 |
+
|
| 40 |
+
### Direct Use
|
| 41 |
+
|
| 42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 43 |
+
|
| 44 |
+
[More Information Needed]
|
| 45 |
+
|
| 46 |
+
### Downstream Use [optional]
|
| 47 |
+
|
| 48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 49 |
+
|
| 50 |
+
[More Information Needed]
|
| 51 |
+
|
| 52 |
+
### Out-of-Scope Use
|
| 53 |
+
|
| 54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 55 |
+
|
| 56 |
+
[More Information Needed]
|
| 57 |
+
|
| 58 |
+
## Bias, Risks, and Limitations
|
| 59 |
+
|
| 60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 61 |
+
|
| 62 |
+
[More Information Needed]
|
| 63 |
+
|
| 64 |
+
### Recommendations
|
| 65 |
+
|
| 66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 67 |
+
|
| 68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 69 |
+
|
| 70 |
+
## How to Get Started with the Model
|
| 71 |
+
|
| 72 |
+
Use the code below to get started with the model.
|
| 73 |
+
|
| 74 |
+
[More Information Needed]
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
|
| 78 |
+
### Training Data
|
| 79 |
+
|
| 80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
### Training Procedure
|
| 85 |
+
|
| 86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 87 |
+
|
| 88 |
+
#### Preprocessing [optional]
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
#### Training Hyperparameters
|
| 94 |
+
|
| 95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 96 |
+
|
| 97 |
+
#### Speeds, Sizes, Times [optional]
|
| 98 |
+
|
| 99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 100 |
+
|
| 101 |
+
[More Information Needed]
|
| 102 |
+
|
| 103 |
+
## Evaluation
|
| 104 |
+
|
| 105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 106 |
+
|
| 107 |
+
### Testing Data, Factors & Metrics
|
| 108 |
+
|
| 109 |
+
#### Testing Data
|
| 110 |
+
|
| 111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 112 |
+
|
| 113 |
+
[More Information Needed]
|
| 114 |
+
|
| 115 |
+
#### Factors
|
| 116 |
+
|
| 117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 118 |
+
|
| 119 |
+
[More Information Needed]
|
| 120 |
+
|
| 121 |
+
#### Metrics
|
| 122 |
+
|
| 123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 124 |
+
|
| 125 |
+
[More Information Needed]
|
| 126 |
+
|
| 127 |
+
### Results
|
| 128 |
+
|
| 129 |
+
[More Information Needed]
|
| 130 |
+
|
| 131 |
+
#### Summary
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
## Model Examination [optional]
|
| 136 |
+
|
| 137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 138 |
+
|
| 139 |
+
[More Information Needed]
|
| 140 |
+
|
| 141 |
+
## Environmental Impact
|
| 142 |
+
|
| 143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 144 |
+
|
| 145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
| 146 |
+
|
| 147 |
+
- **Hardware Type:** [More Information Needed]
|
| 148 |
+
- **Hours used:** [More Information Needed]
|
| 149 |
+
- **Cloud Provider:** [More Information Needed]
|
| 150 |
+
- **Compute Region:** [More Information Needed]
|
| 151 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 152 |
+
|
| 153 |
+
## Technical Specifications [optional]
|
| 154 |
+
|
| 155 |
+
### Model Architecture and Objective
|
| 156 |
+
|
| 157 |
+
[More Information Needed]
|
| 158 |
+
|
| 159 |
+
### Compute Infrastructure
|
| 160 |
+
|
| 161 |
+
[More Information Needed]
|
| 162 |
+
|
| 163 |
+
#### Hardware
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
#### Software
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
## Citation [optional]
|
| 172 |
+
|
| 173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 174 |
+
|
| 175 |
+
**BibTeX:**
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
**APA:**
|
| 180 |
+
|
| 181 |
+
[More Information Needed]
|
| 182 |
+
|
| 183 |
+
## Glossary [optional]
|
| 184 |
+
|
| 185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 186 |
+
|
| 187 |
+
[More Information Needed]
|
| 188 |
+
|
| 189 |
+
## More Information [optional]
|
| 190 |
+
|
| 191 |
+
[More Information Needed]
|
| 192 |
+
|
| 193 |
+
## Model Card Authors [optional]
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## Model Card Contact
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
### Framework versions
|
| 201 |
+
|
| 202 |
+
- PEFT 0.14.0
|
checkpoint-361/adapter_config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "rinna/llama-3-youko-8b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"eva_config": null,
|
| 7 |
+
"exclude_modules": null,
|
| 8 |
+
"fan_in_fan_out": false,
|
| 9 |
+
"inference_mode": true,
|
| 10 |
+
"init_lora_weights": true,
|
| 11 |
+
"layer_replication": null,
|
| 12 |
+
"layers_pattern": null,
|
| 13 |
+
"layers_to_transform": null,
|
| 14 |
+
"loftq_config": {},
|
| 15 |
+
"lora_alpha": 32,
|
| 16 |
+
"lora_bias": false,
|
| 17 |
+
"lora_dropout": 0,
|
| 18 |
+
"megatron_config": null,
|
| 19 |
+
"megatron_core": "megatron.core",
|
| 20 |
+
"modules_to_save": [
|
| 21 |
+
"lm_head",
|
| 22 |
+
"embed_tokens"
|
| 23 |
+
],
|
| 24 |
+
"peft_type": "LORA",
|
| 25 |
+
"r": 128,
|
| 26 |
+
"rank_pattern": {},
|
| 27 |
+
"revision": null,
|
| 28 |
+
"target_modules": [
|
| 29 |
+
"q_proj",
|
| 30 |
+
"o_proj",
|
| 31 |
+
"v_proj",
|
| 32 |
+
"up_proj",
|
| 33 |
+
"k_proj",
|
| 34 |
+
"down_proj",
|
| 35 |
+
"gate_proj"
|
| 36 |
+
],
|
| 37 |
+
"task_type": "CAUSAL_LM",
|
| 38 |
+
"use_dora": false,
|
| 39 |
+
"use_rslora": true
|
| 40 |
+
}
|
checkpoint-361/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c3a6fec0f139dd2f027cae235305ec013753fa9334b4299f338d5995106e660
|
| 3 |
+
size 3443585096
|
checkpoint-361/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bbd79f5146913dd8969f99695c7a4342ed1c753c28053e2e24c183c18b5ef45d
|
| 3 |
+
size 13774624938
|
checkpoint-361/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4c917636c7a58af68a29056522a757e9f9b99005b776641aa157c536967817d
|
| 3 |
+
size 14244
|
checkpoint-361/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9af66bea3f876182eb22d999a5c4cd7c0b4233e355796dc5d83eb977f65c9a2a
|
| 3 |
+
size 1064
|
checkpoint-361/trainer_state.json
ADDED
|
@@ -0,0 +1,2560 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 361,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.002770083102493075,
|
| 13 |
+
"grad_norm": 15.603601455688477,
|
| 14 |
+
"learning_rate": 7.5e-06,
|
| 15 |
+
"loss": 1.4608,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.00554016620498615,
|
| 20 |
+
"grad_norm": 14.760490417480469,
|
| 21 |
+
"learning_rate": 1.5e-05,
|
| 22 |
+
"loss": 1.428,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.008310249307479225,
|
| 27 |
+
"grad_norm": 19.810142517089844,
|
| 28 |
+
"learning_rate": 2.25e-05,
|
| 29 |
+
"loss": 1.2831,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0110803324099723,
|
| 34 |
+
"grad_norm": 6.9140729904174805,
|
| 35 |
+
"learning_rate": 3e-05,
|
| 36 |
+
"loss": 1.0688,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.013850415512465374,
|
| 41 |
+
"grad_norm": 6.4214253425598145,
|
| 42 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 43 |
+
"loss": 1.2067,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01662049861495845,
|
| 48 |
+
"grad_norm": 11.87337589263916,
|
| 49 |
+
"learning_rate": 4.5e-05,
|
| 50 |
+
"loss": 1.4816,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.019390581717451522,
|
| 55 |
+
"grad_norm": 6.669366359710693,
|
| 56 |
+
"learning_rate": 5.25e-05,
|
| 57 |
+
"loss": 1.1287,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0221606648199446,
|
| 62 |
+
"grad_norm": 8.616178512573242,
|
| 63 |
+
"learning_rate": 6e-05,
|
| 64 |
+
"loss": 1.2203,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.024930747922437674,
|
| 69 |
+
"grad_norm": 9.265680313110352,
|
| 70 |
+
"learning_rate": 5.999881193903433e-05,
|
| 71 |
+
"loss": 1.1897,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.027700831024930747,
|
| 76 |
+
"grad_norm": 5.8777852058410645,
|
| 77 |
+
"learning_rate": 5.999524785023657e-05,
|
| 78 |
+
"loss": 1.0788,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.030470914127423823,
|
| 83 |
+
"grad_norm": 5.538949012756348,
|
| 84 |
+
"learning_rate": 5.998930801589704e-05,
|
| 85 |
+
"loss": 1.0736,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0332409972299169,
|
| 90 |
+
"grad_norm": 5.686643123626709,
|
| 91 |
+
"learning_rate": 5.9980992906474764e-05,
|
| 92 |
+
"loss": 1.1573,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.036011080332409975,
|
| 97 |
+
"grad_norm": 6.211599349975586,
|
| 98 |
+
"learning_rate": 5.9970303180560196e-05,
|
| 99 |
+
"loss": 1.1618,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.038781163434903045,
|
| 104 |
+
"grad_norm": 5.079687595367432,
|
| 105 |
+
"learning_rate": 5.995723968482309e-05,
|
| 106 |
+
"loss": 0.9948,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.04155124653739612,
|
| 111 |
+
"grad_norm": 5.150979042053223,
|
| 112 |
+
"learning_rate": 5.994180345394539e-05,
|
| 113 |
+
"loss": 1.0328,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.0443213296398892,
|
| 118 |
+
"grad_norm": 5.891695499420166,
|
| 119 |
+
"learning_rate": 5.9923995710539324e-05,
|
| 120 |
+
"loss": 1.1854,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04709141274238227,
|
| 125 |
+
"grad_norm": 5.33378791809082,
|
| 126 |
+
"learning_rate": 5.990381786505055e-05,
|
| 127 |
+
"loss": 1.0424,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.04986149584487535,
|
| 132 |
+
"grad_norm": 4.58151912689209,
|
| 133 |
+
"learning_rate": 5.988127151564644e-05,
|
| 134 |
+
"loss": 1.0392,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.05263157894736842,
|
| 139 |
+
"grad_norm": 5.48449182510376,
|
| 140 |
+
"learning_rate": 5.9856358448089506e-05,
|
| 141 |
+
"loss": 1.147,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.055401662049861494,
|
| 146 |
+
"grad_norm": 5.061849117279053,
|
| 147 |
+
"learning_rate": 5.9829080635595944e-05,
|
| 148 |
+
"loss": 1.1072,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.05817174515235457,
|
| 153 |
+
"grad_norm": 4.828271389007568,
|
| 154 |
+
"learning_rate": 5.979944023867938e-05,
|
| 155 |
+
"loss": 1.1132,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.060941828254847646,
|
| 160 |
+
"grad_norm": 5.035257816314697,
|
| 161 |
+
"learning_rate": 5.976743960497973e-05,
|
| 162 |
+
"loss": 1.0272,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.06371191135734072,
|
| 167 |
+
"grad_norm": 5.55271577835083,
|
| 168 |
+
"learning_rate": 5.973308126907723e-05,
|
| 169 |
+
"loss": 1.1674,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.0664819944598338,
|
| 174 |
+
"grad_norm": 4.926880836486816,
|
| 175 |
+
"learning_rate": 5.969636795229172e-05,
|
| 176 |
+
"loss": 1.0765,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.06925207756232687,
|
| 181 |
+
"grad_norm": 5.437366962432861,
|
| 182 |
+
"learning_rate": 5.965730256246713e-05,
|
| 183 |
+
"loss": 1.1344,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.07202216066481995,
|
| 188 |
+
"grad_norm": 4.4522786140441895,
|
| 189 |
+
"learning_rate": 5.9615888193741094e-05,
|
| 190 |
+
"loss": 0.9976,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.07479224376731301,
|
| 195 |
+
"grad_norm": 5.204329967498779,
|
| 196 |
+
"learning_rate": 5.9572128126299946e-05,
|
| 197 |
+
"loss": 1.1298,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07756232686980609,
|
| 202 |
+
"grad_norm": 5.264033317565918,
|
| 203 |
+
"learning_rate": 5.9526025826118885e-05,
|
| 204 |
+
"loss": 1.0428,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08033240997229917,
|
| 209 |
+
"grad_norm": 5.672830581665039,
|
| 210 |
+
"learning_rate": 5.947758494468746e-05,
|
| 211 |
+
"loss": 1.1636,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.08310249307479224,
|
| 216 |
+
"grad_norm": 5.798141002655029,
|
| 217 |
+
"learning_rate": 5.942680931872036e-05,
|
| 218 |
+
"loss": 1.1112,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.08587257617728532,
|
| 223 |
+
"grad_norm": 4.760410308837891,
|
| 224 |
+
"learning_rate": 5.937370296985354e-05,
|
| 225 |
+
"loss": 1.0444,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.0886426592797784,
|
| 230 |
+
"grad_norm": 5.4676313400268555,
|
| 231 |
+
"learning_rate": 5.931827010432566e-05,
|
| 232 |
+
"loss": 1.0957,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.09141274238227147,
|
| 237 |
+
"grad_norm": 4.983479976654053,
|
| 238 |
+
"learning_rate": 5.9260515112644995e-05,
|
| 239 |
+
"loss": 1.0734,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.09418282548476455,
|
| 244 |
+
"grad_norm": 5.242843151092529,
|
| 245 |
+
"learning_rate": 5.9200442569241606e-05,
|
| 246 |
+
"loss": 1.0879,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.09695290858725762,
|
| 251 |
+
"grad_norm": 5.12119722366333,
|
| 252 |
+
"learning_rate": 5.9138057232105084e-05,
|
| 253 |
+
"loss": 1.0841,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.0997229916897507,
|
| 258 |
+
"grad_norm": 5.087031841278076,
|
| 259 |
+
"learning_rate": 5.9073364042407705e-05,
|
| 260 |
+
"loss": 1.0343,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.10249307479224377,
|
| 265 |
+
"grad_norm": 5.226377487182617,
|
| 266 |
+
"learning_rate": 5.900636812411301e-05,
|
| 267 |
+
"loss": 1.0775,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.10526315789473684,
|
| 272 |
+
"grad_norm": 5.388728141784668,
|
| 273 |
+
"learning_rate": 5.893707478357005e-05,
|
| 274 |
+
"loss": 1.0439,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.10803324099722991,
|
| 279 |
+
"grad_norm": 5.6382951736450195,
|
| 280 |
+
"learning_rate": 5.886548950909301e-05,
|
| 281 |
+
"loss": 1.096,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.11080332409972299,
|
| 286 |
+
"grad_norm": 5.046523094177246,
|
| 287 |
+
"learning_rate": 5.879161797052658e-05,
|
| 288 |
+
"loss": 0.9928,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.11357340720221606,
|
| 293 |
+
"grad_norm": 5.98545503616333,
|
| 294 |
+
"learning_rate": 5.8715466018796865e-05,
|
| 295 |
+
"loss": 1.1639,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.11634349030470914,
|
| 300 |
+
"grad_norm": 5.064982891082764,
|
| 301 |
+
"learning_rate": 5.863703968544795e-05,
|
| 302 |
+
"loss": 0.9878,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.11911357340720222,
|
| 307 |
+
"grad_norm": 5.36715841293335,
|
| 308 |
+
"learning_rate": 5.8556345182164176e-05,
|
| 309 |
+
"loss": 1.056,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.12188365650969529,
|
| 314 |
+
"grad_norm": 4.7610368728637695,
|
| 315 |
+
"learning_rate": 5.8473388900278206e-05,
|
| 316 |
+
"loss": 0.9625,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.12465373961218837,
|
| 321 |
+
"grad_norm": 5.5188889503479,
|
| 322 |
+
"learning_rate": 5.83881774102647e-05,
|
| 323 |
+
"loss": 1.0828,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.12742382271468145,
|
| 328 |
+
"grad_norm": 5.703125476837158,
|
| 329 |
+
"learning_rate": 5.8300717461220027e-05,
|
| 330 |
+
"loss": 1.0632,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.13019390581717452,
|
| 335 |
+
"grad_norm": 5.264132499694824,
|
| 336 |
+
"learning_rate": 5.821101598032759e-05,
|
| 337 |
+
"loss": 1.0057,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1329639889196676,
|
| 342 |
+
"grad_norm": 6.05692720413208,
|
| 343 |
+
"learning_rate": 5.811908007230929e-05,
|
| 344 |
+
"loss": 1.0565,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.13573407202216067,
|
| 349 |
+
"grad_norm": 4.773650169372559,
|
| 350 |
+
"learning_rate": 5.802491701886268e-05,
|
| 351 |
+
"loss": 0.9764,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.13850415512465375,
|
| 356 |
+
"grad_norm": 5.461302757263184,
|
| 357 |
+
"learning_rate": 5.792853427808431e-05,
|
| 358 |
+
"loss": 1.0002,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.14127423822714683,
|
| 363 |
+
"grad_norm": 5.83071231842041,
|
| 364 |
+
"learning_rate": 5.7829939483878996e-05,
|
| 365 |
+
"loss": 1.0353,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1440443213296399,
|
| 370 |
+
"grad_norm": 5.715704917907715,
|
| 371 |
+
"learning_rate": 5.772914044535516e-05,
|
| 372 |
+
"loss": 0.985,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.14681440443213298,
|
| 377 |
+
"grad_norm": 5.309004306793213,
|
| 378 |
+
"learning_rate": 5.762614514620634e-05,
|
| 379 |
+
"loss": 1.0106,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.14958448753462603,
|
| 384 |
+
"grad_norm": 5.693915843963623,
|
| 385 |
+
"learning_rate": 5.752096174407884e-05,
|
| 386 |
+
"loss": 1.0049,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.1523545706371191,
|
| 391 |
+
"grad_norm": 5.644995212554932,
|
| 392 |
+
"learning_rate": 5.741359856992561e-05,
|
| 393 |
+
"loss": 1.0411,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.15512465373961218,
|
| 398 |
+
"grad_norm": 5.765170574188232,
|
| 399 |
+
"learning_rate": 5.73040641273464e-05,
|
| 400 |
+
"loss": 1.0039,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.15789473684210525,
|
| 405 |
+
"grad_norm": 5.643161296844482,
|
| 406 |
+
"learning_rate": 5.719236709191428e-05,
|
| 407 |
+
"loss": 1.0173,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.16066481994459833,
|
| 412 |
+
"grad_norm": 5.821686267852783,
|
| 413 |
+
"learning_rate": 5.707851631048841e-05,
|
| 414 |
+
"loss": 0.9725,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1634349030470914,
|
| 419 |
+
"grad_norm": 6.034667015075684,
|
| 420 |
+
"learning_rate": 5.6962520800513414e-05,
|
| 421 |
+
"loss": 1.0381,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.16620498614958448,
|
| 426 |
+
"grad_norm": 5.336913585662842,
|
| 427 |
+
"learning_rate": 5.6844389749305136e-05,
|
| 428 |
+
"loss": 0.9266,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.16897506925207756,
|
| 433 |
+
"grad_norm": 5.65543270111084,
|
| 434 |
+
"learning_rate": 5.672413251332297e-05,
|
| 435 |
+
"loss": 0.9768,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.17174515235457063,
|
| 440 |
+
"grad_norm": 5.3824615478515625,
|
| 441 |
+
"learning_rate": 5.6601758617428766e-05,
|
| 442 |
+
"loss": 0.9276,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.1745152354570637,
|
| 447 |
+
"grad_norm": 5.195915222167969,
|
| 448 |
+
"learning_rate": 5.647727775413245e-05,
|
| 449 |
+
"loss": 0.9332,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.1772853185595568,
|
| 454 |
+
"grad_norm": 5.240732669830322,
|
| 455 |
+
"learning_rate": 5.6350699782824346e-05,
|
| 456 |
+
"loss": 0.9012,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.18005540166204986,
|
| 461 |
+
"grad_norm": 5.112541675567627,
|
| 462 |
+
"learning_rate": 5.622203472899423e-05,
|
| 463 |
+
"loss": 0.9211,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.18282548476454294,
|
| 468 |
+
"grad_norm": 6.605805397033691,
|
| 469 |
+
"learning_rate": 5.609129278343731e-05,
|
| 470 |
+
"loss": 0.9847,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.18559556786703602,
|
| 475 |
+
"grad_norm": 5.1862568855285645,
|
| 476 |
+
"learning_rate": 5.595848430144705e-05,
|
| 477 |
+
"loss": 0.9064,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.1883656509695291,
|
| 482 |
+
"grad_norm": 6.03863525390625,
|
| 483 |
+
"learning_rate": 5.582361980199504e-05,
|
| 484 |
+
"loss": 0.9405,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.19113573407202217,
|
| 489 |
+
"grad_norm": 6.4853901863098145,
|
| 490 |
+
"learning_rate": 5.568670996689773e-05,
|
| 491 |
+
"loss": 0.9676,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.19390581717451524,
|
| 496 |
+
"grad_norm": 5.415260314941406,
|
| 497 |
+
"learning_rate": 5.554776563997056e-05,
|
| 498 |
+
"loss": 0.8873,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.19667590027700832,
|
| 503 |
+
"grad_norm": 5.11320686340332,
|
| 504 |
+
"learning_rate": 5.540679782616892e-05,
|
| 505 |
+
"loss": 0.8752,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.1994459833795014,
|
| 510 |
+
"grad_norm": 4.8276214599609375,
|
| 511 |
+
"learning_rate": 5.52638176907166e-05,
|
| 512 |
+
"loss": 0.8441,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.20221606648199447,
|
| 517 |
+
"grad_norm": 5.574111461639404,
|
| 518 |
+
"learning_rate": 5.5118836558221475e-05,
|
| 519 |
+
"loss": 0.907,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.20498614958448755,
|
| 524 |
+
"grad_norm": 5.575868606567383,
|
| 525 |
+
"learning_rate": 5.497186591177849e-05,
|
| 526 |
+
"loss": 0.8691,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.2077562326869806,
|
| 531 |
+
"grad_norm": 5.388404846191406,
|
| 532 |
+
"learning_rate": 5.4822917392060184e-05,
|
| 533 |
+
"loss": 0.8935,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.21052631578947367,
|
| 538 |
+
"grad_norm": 6.2548747062683105,
|
| 539 |
+
"learning_rate": 5.46720027963947e-05,
|
| 540 |
+
"loss": 0.8814,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.21329639889196675,
|
| 545 |
+
"grad_norm": 5.358329772949219,
|
| 546 |
+
"learning_rate": 5.4519134077831395e-05,
|
| 547 |
+
"loss": 0.9041,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.21606648199445982,
|
| 552 |
+
"grad_norm": 6.0121564865112305,
|
| 553 |
+
"learning_rate": 5.4364323344194095e-05,
|
| 554 |
+
"loss": 0.8568,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2188365650969529,
|
| 559 |
+
"grad_norm": 6.2294020652771,
|
| 560 |
+
"learning_rate": 5.420758285712211e-05,
|
| 561 |
+
"loss": 0.9205,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.22160664819944598,
|
| 566 |
+
"grad_norm": 5.889683723449707,
|
| 567 |
+
"learning_rate": 5.404892503109906e-05,
|
| 568 |
+
"loss": 0.8264,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.22437673130193905,
|
| 573 |
+
"grad_norm": 5.401793956756592,
|
| 574 |
+
"learning_rate": 5.388836243246963e-05,
|
| 575 |
+
"loss": 0.8574,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.22714681440443213,
|
| 580 |
+
"grad_norm": 6.194601058959961,
|
| 581 |
+
"learning_rate": 5.372590777844421e-05,
|
| 582 |
+
"loss": 0.8742,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2299168975069252,
|
| 587 |
+
"grad_norm": 5.504939079284668,
|
| 588 |
+
"learning_rate": 5.356157393609167e-05,
|
| 589 |
+
"loss": 0.8261,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.23268698060941828,
|
| 594 |
+
"grad_norm": 5.818756103515625,
|
| 595 |
+
"learning_rate": 5.339537392132025e-05,
|
| 596 |
+
"loss": 0.8364,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.23545706371191136,
|
| 601 |
+
"grad_norm": 5.8196024894714355,
|
| 602 |
+
"learning_rate": 5.322732089784661e-05,
|
| 603 |
+
"loss": 0.8577,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.23822714681440443,
|
| 608 |
+
"grad_norm": 6.102773189544678,
|
| 609 |
+
"learning_rate": 5.305742817615325e-05,
|
| 610 |
+
"loss": 0.8708,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2409972299168975,
|
| 615 |
+
"grad_norm": 6.225861549377441,
|
| 616 |
+
"learning_rate": 5.288570921243423e-05,
|
| 617 |
+
"loss": 0.8654,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.24376731301939059,
|
| 622 |
+
"grad_norm": 6.0107340812683105,
|
| 623 |
+
"learning_rate": 5.2712177607529405e-05,
|
| 624 |
+
"loss": 0.8401,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.24653739612188366,
|
| 629 |
+
"grad_norm": 6.3431715965271,
|
| 630 |
+
"learning_rate": 5.2536847105847185e-05,
|
| 631 |
+
"loss": 0.855,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.24930747922437674,
|
| 636 |
+
"grad_norm": 5.196923732757568,
|
| 637 |
+
"learning_rate": 5.235973159427591e-05,
|
| 638 |
+
"loss": 0.756,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2520775623268698,
|
| 643 |
+
"grad_norm": 6.157230377197266,
|
| 644 |
+
"learning_rate": 5.218084510108397e-05,
|
| 645 |
+
"loss": 0.856,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2548476454293629,
|
| 650 |
+
"grad_norm": 5.813492774963379,
|
| 651 |
+
"learning_rate": 5.200020179480868e-05,
|
| 652 |
+
"loss": 0.7829,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.25761772853185594,
|
| 657 |
+
"grad_norm": 6.468319416046143,
|
| 658 |
+
"learning_rate": 5.181781598313409e-05,
|
| 659 |
+
"loss": 0.8327,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.26038781163434904,
|
| 664 |
+
"grad_norm": 6.1590423583984375,
|
| 665 |
+
"learning_rate": 5.16337021117578e-05,
|
| 666 |
+
"loss": 0.8035,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.2631578947368421,
|
| 671 |
+
"grad_norm": 6.378724575042725,
|
| 672 |
+
"learning_rate": 5.14478747632467e-05,
|
| 673 |
+
"loss": 0.8134,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.2659279778393352,
|
| 678 |
+
"grad_norm": 5.47311544418335,
|
| 679 |
+
"learning_rate": 5.126034865588208e-05,
|
| 680 |
+
"loss": 0.7149,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.26869806094182824,
|
| 685 |
+
"grad_norm": 5.790994644165039,
|
| 686 |
+
"learning_rate": 5.107113864249381e-05,
|
| 687 |
+
"loss": 0.7714,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.27146814404432135,
|
| 692 |
+
"grad_norm": 5.880769729614258,
|
| 693 |
+
"learning_rate": 5.088025970928399e-05,
|
| 694 |
+
"loss": 0.7519,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2742382271468144,
|
| 699 |
+
"grad_norm": 5.894231796264648,
|
| 700 |
+
"learning_rate": 5.068772697463992e-05,
|
| 701 |
+
"loss": 0.8124,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2770083102493075,
|
| 706 |
+
"grad_norm": 6.011329174041748,
|
| 707 |
+
"learning_rate": 5.0493555687936704e-05,
|
| 708 |
+
"loss": 0.7536,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.27977839335180055,
|
| 713 |
+
"grad_norm": 5.813382148742676,
|
| 714 |
+
"learning_rate": 5.029776122832945e-05,
|
| 715 |
+
"loss": 0.7483,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.28254847645429365,
|
| 720 |
+
"grad_norm": 6.445542335510254,
|
| 721 |
+
"learning_rate": 5.0100359103535134e-05,
|
| 722 |
+
"loss": 0.7376,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.2853185595567867,
|
| 727 |
+
"grad_norm": 5.571536540985107,
|
| 728 |
+
"learning_rate": 4.9901364948604366e-05,
|
| 729 |
+
"loss": 0.7538,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.2880886426592798,
|
| 734 |
+
"grad_norm": 5.566730499267578,
|
| 735 |
+
"learning_rate": 4.970079452468298e-05,
|
| 736 |
+
"loss": 0.7102,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.29085872576177285,
|
| 741 |
+
"grad_norm": 6.336560249328613,
|
| 742 |
+
"learning_rate": 4.949866371776376e-05,
|
| 743 |
+
"loss": 0.7469,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.29362880886426596,
|
| 748 |
+
"grad_norm": 5.440810203552246,
|
| 749 |
+
"learning_rate": 4.929498853742816e-05,
|
| 750 |
+
"loss": 0.7252,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.296398891966759,
|
| 755 |
+
"grad_norm": 6.162292957305908,
|
| 756 |
+
"learning_rate": 4.908978511557827e-05,
|
| 757 |
+
"loss": 0.7569,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.29916897506925205,
|
| 762 |
+
"grad_norm": 6.059966087341309,
|
| 763 |
+
"learning_rate": 4.8883069705159104e-05,
|
| 764 |
+
"loss": 0.7136,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.30193905817174516,
|
| 769 |
+
"grad_norm": 6.249420642852783,
|
| 770 |
+
"learning_rate": 4.867485867887136e-05,
|
| 771 |
+
"loss": 0.7408,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3047091412742382,
|
| 776 |
+
"grad_norm": 5.173600673675537,
|
| 777 |
+
"learning_rate": 4.846516852787457e-05,
|
| 778 |
+
"loss": 0.7106,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3074792243767313,
|
| 783 |
+
"grad_norm": 5.633162975311279,
|
| 784 |
+
"learning_rate": 4.8254015860480934e-05,
|
| 785 |
+
"loss": 0.6951,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.31024930747922436,
|
| 790 |
+
"grad_norm": 5.940577507019043,
|
| 791 |
+
"learning_rate": 4.804141740083993e-05,
|
| 792 |
+
"loss": 0.7022,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.31301939058171746,
|
| 797 |
+
"grad_norm": 5.466473579406738,
|
| 798 |
+
"learning_rate": 4.7827389987613635e-05,
|
| 799 |
+
"loss": 0.6694,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3157894736842105,
|
| 804 |
+
"grad_norm": 6.02973747253418,
|
| 805 |
+
"learning_rate": 4.761195057264307e-05,
|
| 806 |
+
"loss": 0.6799,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3185595567867036,
|
| 811 |
+
"grad_norm": 5.916580677032471,
|
| 812 |
+
"learning_rate": 4.739511621960551e-05,
|
| 813 |
+
"loss": 0.7019,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.32132963988919666,
|
| 818 |
+
"grad_norm": 5.600930213928223,
|
| 819 |
+
"learning_rate": 4.717690410266304e-05,
|
| 820 |
+
"loss": 0.6818,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.32409972299168976,
|
| 825 |
+
"grad_norm": 5.327589988708496,
|
| 826 |
+
"learning_rate": 4.69573315051022e-05,
|
| 827 |
+
"loss": 0.6811,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3268698060941828,
|
| 832 |
+
"grad_norm": 5.755538463592529,
|
| 833 |
+
"learning_rate": 4.673641581796515e-05,
|
| 834 |
+
"loss": 0.6705,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3296398891966759,
|
| 839 |
+
"grad_norm": 6.217751979827881,
|
| 840 |
+
"learning_rate": 4.6514174538672187e-05,
|
| 841 |
+
"loss": 0.6856,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.33240997229916897,
|
| 846 |
+
"grad_norm": 5.41038179397583,
|
| 847 |
+
"learning_rate": 4.629062526963592e-05,
|
| 848 |
+
"loss": 0.6509,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.33518005540166207,
|
| 853 |
+
"grad_norm": 5.7826247215271,
|
| 854 |
+
"learning_rate": 4.606578571686703e-05,
|
| 855 |
+
"loss": 0.662,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.3379501385041551,
|
| 860 |
+
"grad_norm": 6.125760555267334,
|
| 861 |
+
"learning_rate": 4.583967368857194e-05,
|
| 862 |
+
"loss": 0.6518,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.3407202216066482,
|
| 867 |
+
"grad_norm": 5.408969879150391,
|
| 868 |
+
"learning_rate": 4.561230709374229e-05,
|
| 869 |
+
"loss": 0.6391,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.34349030470914127,
|
| 874 |
+
"grad_norm": 5.880269527435303,
|
| 875 |
+
"learning_rate": 4.5383703940736494e-05,
|
| 876 |
+
"loss": 0.6388,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.3462603878116344,
|
| 881 |
+
"grad_norm": 6.065842628479004,
|
| 882 |
+
"learning_rate": 4.51538823358534e-05,
|
| 883 |
+
"loss": 0.653,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.3490304709141274,
|
| 888 |
+
"grad_norm": 6.418548107147217,
|
| 889 |
+
"learning_rate": 4.49228604818982e-05,
|
| 890 |
+
"loss": 0.6607,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3518005540166205,
|
| 895 |
+
"grad_norm": 5.7509074211120605,
|
| 896 |
+
"learning_rate": 4.469065667674069e-05,
|
| 897 |
+
"loss": 0.6289,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3545706371191136,
|
| 902 |
+
"grad_norm": 5.758950710296631,
|
| 903 |
+
"learning_rate": 4.445728931186599e-05,
|
| 904 |
+
"loss": 0.6332,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3573407202216066,
|
| 909 |
+
"grad_norm": 5.907932758331299,
|
| 910 |
+
"learning_rate": 4.42227768709179e-05,
|
| 911 |
+
"loss": 0.634,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.3601108033240997,
|
| 916 |
+
"grad_norm": 5.904737949371338,
|
| 917 |
+
"learning_rate": 4.398713792823489e-05,
|
| 918 |
+
"loss": 0.5948,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3628808864265928,
|
| 923 |
+
"grad_norm": 6.62101411819458,
|
| 924 |
+
"learning_rate": 4.375039114737895e-05,
|
| 925 |
+
"loss": 0.6332,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3656509695290859,
|
| 930 |
+
"grad_norm": 6.03896951675415,
|
| 931 |
+
"learning_rate": 4.3512555279657344e-05,
|
| 932 |
+
"loss": 0.6153,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.3684210526315789,
|
| 937 |
+
"grad_norm": 5.788692474365234,
|
| 938 |
+
"learning_rate": 4.3273649162637454e-05,
|
| 939 |
+
"loss": 0.6223,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.37119113573407203,
|
| 944 |
+
"grad_norm": 5.691258907318115,
|
| 945 |
+
"learning_rate": 4.3033691718654767e-05,
|
| 946 |
+
"loss": 0.6048,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.3739612188365651,
|
| 951 |
+
"grad_norm": 5.318130016326904,
|
| 952 |
+
"learning_rate": 4.2792701953314085e-05,
|
| 953 |
+
"loss": 0.6023,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.3767313019390582,
|
| 958 |
+
"grad_norm": 6.409132480621338,
|
| 959 |
+
"learning_rate": 4.2550698953984335e-05,
|
| 960 |
+
"loss": 0.5975,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.37950138504155123,
|
| 965 |
+
"grad_norm": 5.915623664855957,
|
| 966 |
+
"learning_rate": 4.230770188828665e-05,
|
| 967 |
+
"loss": 0.6048,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.38227146814404434,
|
| 972 |
+
"grad_norm": 6.316535472869873,
|
| 973 |
+
"learning_rate": 4.2063730002576244e-05,
|
| 974 |
+
"loss": 0.5944,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.3850415512465374,
|
| 979 |
+
"grad_norm": 6.343788146972656,
|
| 980 |
+
"learning_rate": 4.181880262041808e-05,
|
| 981 |
+
"loss": 0.6115,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.3878116343490305,
|
| 986 |
+
"grad_norm": 6.030927658081055,
|
| 987 |
+
"learning_rate": 4.15729391410563e-05,
|
| 988 |
+
"loss": 0.5709,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.39058171745152354,
|
| 993 |
+
"grad_norm": 5.820158004760742,
|
| 994 |
+
"learning_rate": 4.132615903787774e-05,
|
| 995 |
+
"loss": 0.5882,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.39335180055401664,
|
| 1000 |
+
"grad_norm": 5.914207935333252,
|
| 1001 |
+
"learning_rate": 4.1078481856869586e-05,
|
| 1002 |
+
"loss": 0.5579,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.3961218836565097,
|
| 1007 |
+
"grad_norm": 5.729140758514404,
|
| 1008 |
+
"learning_rate": 4.0829927215071234e-05,
|
| 1009 |
+
"loss": 0.569,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.3988919667590028,
|
| 1014 |
+
"grad_norm": 6.345619201660156,
|
| 1015 |
+
"learning_rate": 4.0580514799020525e-05,
|
| 1016 |
+
"loss": 0.5574,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.40166204986149584,
|
| 1021 |
+
"grad_norm": 5.319207668304443,
|
| 1022 |
+
"learning_rate": 4.033026436319452e-05,
|
| 1023 |
+
"loss": 0.5584,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.40443213296398894,
|
| 1028 |
+
"grad_norm": 6.186460018157959,
|
| 1029 |
+
"learning_rate": 4.0079195728444845e-05,
|
| 1030 |
+
"loss": 0.5476,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.407202216066482,
|
| 1035 |
+
"grad_norm": 6.213712692260742,
|
| 1036 |
+
"learning_rate": 3.982732878042782e-05,
|
| 1037 |
+
"loss": 0.5403,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.4099722991689751,
|
| 1042 |
+
"grad_norm": 5.564090251922607,
|
| 1043 |
+
"learning_rate": 3.95746834680294e-05,
|
| 1044 |
+
"loss": 0.5391,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.41274238227146814,
|
| 1049 |
+
"grad_norm": 6.060794830322266,
|
| 1050 |
+
"learning_rate": 3.932127980178519e-05,
|
| 1051 |
+
"loss": 0.5332,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.4155124653739612,
|
| 1056 |
+
"grad_norm": 5.621760368347168,
|
| 1057 |
+
"learning_rate": 3.906713785229546e-05,
|
| 1058 |
+
"loss": 0.5426,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.4182825484764543,
|
| 1063 |
+
"grad_norm": 5.4746012687683105,
|
| 1064 |
+
"learning_rate": 3.8812277748635554e-05,
|
| 1065 |
+
"loss": 0.5444,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.42105263157894735,
|
| 1070 |
+
"grad_norm": 4.968982696533203,
|
| 1071 |
+
"learning_rate": 3.855671967676153e-05,
|
| 1072 |
+
"loss": 0.525,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.42382271468144045,
|
| 1077 |
+
"grad_norm": 5.877557277679443,
|
| 1078 |
+
"learning_rate": 3.830048387791136e-05,
|
| 1079 |
+
"loss": 0.5485,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4265927977839335,
|
| 1084 |
+
"grad_norm": 5.59464168548584,
|
| 1085 |
+
"learning_rate": 3.804359064700175e-05,
|
| 1086 |
+
"loss": 0.5274,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4293628808864266,
|
| 1091 |
+
"grad_norm": 5.718829154968262,
|
| 1092 |
+
"learning_rate": 3.778606033102071e-05,
|
| 1093 |
+
"loss": 0.5321,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.43213296398891965,
|
| 1098 |
+
"grad_norm": 5.996484279632568,
|
| 1099 |
+
"learning_rate": 3.752791332741596e-05,
|
| 1100 |
+
"loss": 0.5083,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.43490304709141275,
|
| 1105 |
+
"grad_norm": 5.6276373863220215,
|
| 1106 |
+
"learning_rate": 3.7269170082479376e-05,
|
| 1107 |
+
"loss": 0.5466,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4376731301939058,
|
| 1112 |
+
"grad_norm": 5.630644798278809,
|
| 1113 |
+
"learning_rate": 3.7009851089727615e-05,
|
| 1114 |
+
"loss": 0.5168,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4404432132963989,
|
| 1119 |
+
"grad_norm": 5.379649639129639,
|
| 1120 |
+
"learning_rate": 3.6749976888278864e-05,
|
| 1121 |
+
"loss": 0.529,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.44321329639889195,
|
| 1126 |
+
"grad_norm": 5.533950328826904,
|
| 1127 |
+
"learning_rate": 3.6489568061226106e-05,
|
| 1128 |
+
"loss": 0.4908,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.44598337950138506,
|
| 1133 |
+
"grad_norm": 6.063986301422119,
|
| 1134 |
+
"learning_rate": 3.6228645234006835e-05,
|
| 1135 |
+
"loss": 0.5209,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4487534626038781,
|
| 1140 |
+
"grad_norm": 6.18126916885376,
|
| 1141 |
+
"learning_rate": 3.596722907276946e-05,
|
| 1142 |
+
"loss": 0.4909,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.4515235457063712,
|
| 1147 |
+
"grad_norm": 5.693446636199951,
|
| 1148 |
+
"learning_rate": 3.570534028273645e-05,
|
| 1149 |
+
"loss": 0.5257,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.45429362880886426,
|
| 1154 |
+
"grad_norm": 4.973173141479492,
|
| 1155 |
+
"learning_rate": 3.5442999606564376e-05,
|
| 1156 |
+
"loss": 0.4936,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.45706371191135736,
|
| 1161 |
+
"grad_norm": 5.8764729499816895,
|
| 1162 |
+
"learning_rate": 3.518022782270106e-05,
|
| 1163 |
+
"loss": 0.4793,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4598337950138504,
|
| 1168 |
+
"grad_norm": 4.693322658538818,
|
| 1169 |
+
"learning_rate": 3.491704574373978e-05,
|
| 1170 |
+
"loss": 0.4836,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.4626038781163435,
|
| 1175 |
+
"grad_norm": 5.623289108276367,
|
| 1176 |
+
"learning_rate": 3.465347421477087e-05,
|
| 1177 |
+
"loss": 0.4814,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.46537396121883656,
|
| 1182 |
+
"grad_norm": 5.063323497772217,
|
| 1183 |
+
"learning_rate": 3.438953411173066e-05,
|
| 1184 |
+
"loss": 0.4801,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.46814404432132967,
|
| 1189 |
+
"grad_norm": 5.001865863800049,
|
| 1190 |
+
"learning_rate": 3.412524633974808e-05,
|
| 1191 |
+
"loss": 0.5035,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.4709141274238227,
|
| 1196 |
+
"grad_norm": 4.878357410430908,
|
| 1197 |
+
"learning_rate": 3.386063183148883e-05,
|
| 1198 |
+
"loss": 0.4814,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.47368421052631576,
|
| 1203 |
+
"grad_norm": 6.276041507720947,
|
| 1204 |
+
"learning_rate": 3.359571154549747e-05,
|
| 1205 |
+
"loss": 0.4853,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.47645429362880887,
|
| 1210 |
+
"grad_norm": 5.440047740936279,
|
| 1211 |
+
"learning_rate": 3.3330506464537356e-05,
|
| 1212 |
+
"loss": 0.4837,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.4792243767313019,
|
| 1217 |
+
"grad_norm": 4.922585964202881,
|
| 1218 |
+
"learning_rate": 3.306503759392882e-05,
|
| 1219 |
+
"loss": 0.4957,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.481994459833795,
|
| 1224 |
+
"grad_norm": 4.886894226074219,
|
| 1225 |
+
"learning_rate": 3.279932595988538e-05,
|
| 1226 |
+
"loss": 0.4831,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.48476454293628807,
|
| 1231 |
+
"grad_norm": 4.792039394378662,
|
| 1232 |
+
"learning_rate": 3.253339260784838e-05,
|
| 1233 |
+
"loss": 0.5031,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.48753462603878117,
|
| 1238 |
+
"grad_norm": 4.9557647705078125,
|
| 1239 |
+
"learning_rate": 3.226725860082018e-05,
|
| 1240 |
+
"loss": 0.4812,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.4903047091412742,
|
| 1245 |
+
"grad_norm": 5.699478626251221,
|
| 1246 |
+
"learning_rate": 3.200094501769581e-05,
|
| 1247 |
+
"loss": 0.4898,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.4930747922437673,
|
| 1252 |
+
"grad_norm": 4.814623832702637,
|
| 1253 |
+
"learning_rate": 3.173447295159344e-05,
|
| 1254 |
+
"loss": 0.4469,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.49584487534626037,
|
| 1259 |
+
"grad_norm": 5.314681053161621,
|
| 1260 |
+
"learning_rate": 3.1467863508183735e-05,
|
| 1261 |
+
"loss": 0.4889,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.4986149584487535,
|
| 1266 |
+
"grad_norm": 5.161649703979492,
|
| 1267 |
+
"learning_rate": 3.1201137804018246e-05,
|
| 1268 |
+
"loss": 0.4543,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5013850415512465,
|
| 1273 |
+
"grad_norm": 5.343588829040527,
|
| 1274 |
+
"learning_rate": 3.093431696485679e-05,
|
| 1275 |
+
"loss": 0.4557,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5041551246537396,
|
| 1280 |
+
"grad_norm": 4.8219499588012695,
|
| 1281 |
+
"learning_rate": 3.0667422123994304e-05,
|
| 1282 |
+
"loss": 0.4607,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5069252077562327,
|
| 1287 |
+
"grad_norm": 5.002239227294922,
|
| 1288 |
+
"learning_rate": 3.040047442058694e-05,
|
| 1289 |
+
"loss": 0.4693,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5096952908587258,
|
| 1294 |
+
"grad_norm": 4.721946716308594,
|
| 1295 |
+
"learning_rate": 3.0133494997977792e-05,
|
| 1296 |
+
"loss": 0.4445,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5124653739612188,
|
| 1301 |
+
"grad_norm": 4.8258538246154785,
|
| 1302 |
+
"learning_rate": 2.9866505002022223e-05,
|
| 1303 |
+
"loss": 0.4712,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5152354570637119,
|
| 1308 |
+
"grad_norm": 5.217929840087891,
|
| 1309 |
+
"learning_rate": 2.9599525579413066e-05,
|
| 1310 |
+
"loss": 0.4282,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.518005540166205,
|
| 1315 |
+
"grad_norm": 5.506514072418213,
|
| 1316 |
+
"learning_rate": 2.9332577876005707e-05,
|
| 1317 |
+
"loss": 0.4733,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5207756232686981,
|
| 1322 |
+
"grad_norm": 4.479320049285889,
|
| 1323 |
+
"learning_rate": 2.9065683035143213e-05,
|
| 1324 |
+
"loss": 0.4892,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5235457063711911,
|
| 1329 |
+
"grad_norm": 4.9687089920043945,
|
| 1330 |
+
"learning_rate": 2.8798862195981763e-05,
|
| 1331 |
+
"loss": 0.4549,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5263157894736842,
|
| 1336 |
+
"grad_norm": 5.03638219833374,
|
| 1337 |
+
"learning_rate": 2.8532136491816266e-05,
|
| 1338 |
+
"loss": 0.4461,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5290858725761773,
|
| 1343 |
+
"grad_norm": 5.558093547821045,
|
| 1344 |
+
"learning_rate": 2.8265527048406564e-05,
|
| 1345 |
+
"loss": 0.45,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5318559556786704,
|
| 1350 |
+
"grad_norm": 4.820427417755127,
|
| 1351 |
+
"learning_rate": 2.7999054982304192e-05,
|
| 1352 |
+
"loss": 0.4567,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5346260387811634,
|
| 1357 |
+
"grad_norm": 4.981771945953369,
|
| 1358 |
+
"learning_rate": 2.7732741399179813e-05,
|
| 1359 |
+
"loss": 0.4754,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5373961218836565,
|
| 1364 |
+
"grad_norm": 4.8197245597839355,
|
| 1365 |
+
"learning_rate": 2.746660739215162e-05,
|
| 1366 |
+
"loss": 0.4623,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5401662049861495,
|
| 1371 |
+
"grad_norm": 4.506078243255615,
|
| 1372 |
+
"learning_rate": 2.7200674040114627e-05,
|
| 1373 |
+
"loss": 0.4579,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5429362880886427,
|
| 1378 |
+
"grad_norm": 5.280223369598389,
|
| 1379 |
+
"learning_rate": 2.693496240607119e-05,
|
| 1380 |
+
"loss": 0.4455,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5457063711911357,
|
| 1385 |
+
"grad_norm": 4.943070411682129,
|
| 1386 |
+
"learning_rate": 2.666949353546265e-05,
|
| 1387 |
+
"loss": 0.4516,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5484764542936288,
|
| 1392 |
+
"grad_norm": 4.689980983734131,
|
| 1393 |
+
"learning_rate": 2.6404288454502546e-05,
|
| 1394 |
+
"loss": 0.4353,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5512465373961218,
|
| 1399 |
+
"grad_norm": 4.386908531188965,
|
| 1400 |
+
"learning_rate": 2.6139368168511176e-05,
|
| 1401 |
+
"loss": 0.47,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.554016620498615,
|
| 1406 |
+
"grad_norm": 4.444925785064697,
|
| 1407 |
+
"learning_rate": 2.5874753660251916e-05,
|
| 1408 |
+
"loss": 0.4538,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.556786703601108,
|
| 1413 |
+
"grad_norm": 4.536672592163086,
|
| 1414 |
+
"learning_rate": 2.561046588826934e-05,
|
| 1415 |
+
"loss": 0.4416,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5595567867036011,
|
| 1420 |
+
"grad_norm": 4.070997714996338,
|
| 1421 |
+
"learning_rate": 2.5346525785229134e-05,
|
| 1422 |
+
"loss": 0.4422,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.5623268698060941,
|
| 1427 |
+
"grad_norm": 4.6580281257629395,
|
| 1428 |
+
"learning_rate": 2.5082954256260227e-05,
|
| 1429 |
+
"loss": 0.4526,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5650969529085873,
|
| 1434 |
+
"grad_norm": 4.6260247230529785,
|
| 1435 |
+
"learning_rate": 2.481977217729894e-05,
|
| 1436 |
+
"loss": 0.4232,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5678670360110804,
|
| 1441 |
+
"grad_norm": 4.411331653594971,
|
| 1442 |
+
"learning_rate": 2.4557000393435636e-05,
|
| 1443 |
+
"loss": 0.4599,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5706371191135734,
|
| 1448 |
+
"grad_norm": 4.103204250335693,
|
| 1449 |
+
"learning_rate": 2.429465971726356e-05,
|
| 1450 |
+
"loss": 0.4349,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5734072022160664,
|
| 1455 |
+
"grad_norm": 4.877063274383545,
|
| 1456 |
+
"learning_rate": 2.403277092723055e-05,
|
| 1457 |
+
"loss": 0.4521,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5761772853185596,
|
| 1462 |
+
"grad_norm": 4.473416328430176,
|
| 1463 |
+
"learning_rate": 2.3771354765993177e-05,
|
| 1464 |
+
"loss": 0.4398,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.5789473684210527,
|
| 1469 |
+
"grad_norm": 4.630215644836426,
|
| 1470 |
+
"learning_rate": 2.351043193877391e-05,
|
| 1471 |
+
"loss": 0.4708,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.5817174515235457,
|
| 1476 |
+
"grad_norm": 4.25671911239624,
|
| 1477 |
+
"learning_rate": 2.3250023111721137e-05,
|
| 1478 |
+
"loss": 0.4167,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5844875346260388,
|
| 1483 |
+
"grad_norm": 4.418769836425781,
|
| 1484 |
+
"learning_rate": 2.2990148910272383e-05,
|
| 1485 |
+
"loss": 0.4562,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5872576177285319,
|
| 1490 |
+
"grad_norm": 4.783448219299316,
|
| 1491 |
+
"learning_rate": 2.273082991752063e-05,
|
| 1492 |
+
"loss": 0.4371,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.590027700831025,
|
| 1497 |
+
"grad_norm": 4.563108921051025,
|
| 1498 |
+
"learning_rate": 2.2472086672584048e-05,
|
| 1499 |
+
"loss": 0.4335,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.592797783933518,
|
| 1504 |
+
"grad_norm": 4.505840301513672,
|
| 1505 |
+
"learning_rate": 2.22139396689793e-05,
|
| 1506 |
+
"loss": 0.4421,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5955678670360111,
|
| 1511 |
+
"grad_norm": 3.867152452468872,
|
| 1512 |
+
"learning_rate": 2.1956409352998253e-05,
|
| 1513 |
+
"loss": 0.479,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5983379501385041,
|
| 1518 |
+
"grad_norm": 4.328498363494873,
|
| 1519 |
+
"learning_rate": 2.169951612208865e-05,
|
| 1520 |
+
"loss": 0.4359,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.6011080332409973,
|
| 1525 |
+
"grad_norm": 4.354103088378906,
|
| 1526 |
+
"learning_rate": 2.144328032323848e-05,
|
| 1527 |
+
"loss": 0.477,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.6038781163434903,
|
| 1532 |
+
"grad_norm": 4.329044342041016,
|
| 1533 |
+
"learning_rate": 2.1187722251364455e-05,
|
| 1534 |
+
"loss": 0.4553,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6066481994459834,
|
| 1539 |
+
"grad_norm": 4.295444488525391,
|
| 1540 |
+
"learning_rate": 2.093286214770454e-05,
|
| 1541 |
+
"loss": 0.4762,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.6094182825484764,
|
| 1546 |
+
"grad_norm": 3.934124231338501,
|
| 1547 |
+
"learning_rate": 2.0678720198214814e-05,
|
| 1548 |
+
"loss": 0.4415,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.6121883656509696,
|
| 1553 |
+
"grad_norm": 4.6317973136901855,
|
| 1554 |
+
"learning_rate": 2.04253165319706e-05,
|
| 1555 |
+
"loss": 0.4382,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.6149584487534626,
|
| 1560 |
+
"grad_norm": 4.10235071182251,
|
| 1561 |
+
"learning_rate": 2.0172671219572175e-05,
|
| 1562 |
+
"loss": 0.438,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.6177285318559557,
|
| 1567 |
+
"grad_norm": 4.445311546325684,
|
| 1568 |
+
"learning_rate": 1.9920804271555153e-05,
|
| 1569 |
+
"loss": 0.4442,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.6204986149584487,
|
| 1574 |
+
"grad_norm": 4.412539958953857,
|
| 1575 |
+
"learning_rate": 1.9669735636805483e-05,
|
| 1576 |
+
"loss": 0.4343,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.6232686980609419,
|
| 1581 |
+
"grad_norm": 4.816469669342041,
|
| 1582 |
+
"learning_rate": 1.941948520097948e-05,
|
| 1583 |
+
"loss": 0.4482,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6260387811634349,
|
| 1588 |
+
"grad_norm": 4.047571182250977,
|
| 1589 |
+
"learning_rate": 1.9170072784928764e-05,
|
| 1590 |
+
"loss": 0.4466,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.628808864265928,
|
| 1595 |
+
"grad_norm": 4.237831115722656,
|
| 1596 |
+
"learning_rate": 1.8921518143130422e-05,
|
| 1597 |
+
"loss": 0.4543,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.631578947368421,
|
| 1602 |
+
"grad_norm": 3.3804972171783447,
|
| 1603 |
+
"learning_rate": 1.867384096212226e-05,
|
| 1604 |
+
"loss": 0.4177,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6343490304709142,
|
| 1609 |
+
"grad_norm": 4.166910648345947,
|
| 1610 |
+
"learning_rate": 1.842706085894371e-05,
|
| 1611 |
+
"loss": 0.4292,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6371191135734072,
|
| 1616 |
+
"grad_norm": 4.192974090576172,
|
| 1617 |
+
"learning_rate": 1.818119737958192e-05,
|
| 1618 |
+
"loss": 0.4366,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6398891966759003,
|
| 1623 |
+
"grad_norm": 4.138956069946289,
|
| 1624 |
+
"learning_rate": 1.7936269997423754e-05,
|
| 1625 |
+
"loss": 0.4494,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6426592797783933,
|
| 1630 |
+
"grad_norm": 3.8270256519317627,
|
| 1631 |
+
"learning_rate": 1.7692298111713357e-05,
|
| 1632 |
+
"loss": 0.4416,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6454293628808865,
|
| 1637 |
+
"grad_norm": 4.01252555847168,
|
| 1638 |
+
"learning_rate": 1.744930104601566e-05,
|
| 1639 |
+
"loss": 0.4533,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6481994459833795,
|
| 1644 |
+
"grad_norm": 4.926167964935303,
|
| 1645 |
+
"learning_rate": 1.7207298046685913e-05,
|
| 1646 |
+
"loss": 0.446,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6509695290858726,
|
| 1651 |
+
"grad_norm": 4.479677200317383,
|
| 1652 |
+
"learning_rate": 1.696630828134525e-05,
|
| 1653 |
+
"loss": 0.4389,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6537396121883656,
|
| 1658 |
+
"grad_norm": 4.288498878479004,
|
| 1659 |
+
"learning_rate": 1.672635083736255e-05,
|
| 1660 |
+
"loss": 0.4411,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6565096952908587,
|
| 1665 |
+
"grad_norm": 3.79034423828125,
|
| 1666 |
+
"learning_rate": 1.6487444720342657e-05,
|
| 1667 |
+
"loss": 0.4244,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6592797783933518,
|
| 1672 |
+
"grad_norm": 4.163748264312744,
|
| 1673 |
+
"learning_rate": 1.6249608852621057e-05,
|
| 1674 |
+
"loss": 0.4365,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6620498614958449,
|
| 1679 |
+
"grad_norm": 4.326589584350586,
|
| 1680 |
+
"learning_rate": 1.601286207176511e-05,
|
| 1681 |
+
"loss": 0.4338,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6648199445983379,
|
| 1686 |
+
"grad_norm": 4.408019065856934,
|
| 1687 |
+
"learning_rate": 1.5777223129082104e-05,
|
| 1688 |
+
"loss": 0.4274,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.667590027700831,
|
| 1693 |
+
"grad_norm": 4.198231220245361,
|
| 1694 |
+
"learning_rate": 1.554271068813402e-05,
|
| 1695 |
+
"loss": 0.4363,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6703601108033241,
|
| 1700 |
+
"grad_norm": 4.108282566070557,
|
| 1701 |
+
"learning_rate": 1.5309343323259316e-05,
|
| 1702 |
+
"loss": 0.4254,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6731301939058172,
|
| 1707 |
+
"grad_norm": 4.349982738494873,
|
| 1708 |
+
"learning_rate": 1.5077139518101804e-05,
|
| 1709 |
+
"loss": 0.4222,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6759002770083102,
|
| 1714 |
+
"grad_norm": 3.9607222080230713,
|
| 1715 |
+
"learning_rate": 1.4846117664146597e-05,
|
| 1716 |
+
"loss": 0.426,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6786703601108033,
|
| 1721 |
+
"grad_norm": 4.317805290222168,
|
| 1722 |
+
"learning_rate": 1.461629605926351e-05,
|
| 1723 |
+
"loss": 0.4356,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6814404432132964,
|
| 1728 |
+
"grad_norm": 4.254571437835693,
|
| 1729 |
+
"learning_rate": 1.4387692906257705e-05,
|
| 1730 |
+
"loss": 0.4234,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6842105263157895,
|
| 1735 |
+
"grad_norm": 4.319187641143799,
|
| 1736 |
+
"learning_rate": 1.4160326311428072e-05,
|
| 1737 |
+
"loss": 0.419,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6869806094182825,
|
| 1742 |
+
"grad_norm": 4.034308910369873,
|
| 1743 |
+
"learning_rate": 1.3934214283132973e-05,
|
| 1744 |
+
"loss": 0.4349,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6897506925207756,
|
| 1749 |
+
"grad_norm": 4.316988468170166,
|
| 1750 |
+
"learning_rate": 1.3709374730364091e-05,
|
| 1751 |
+
"loss": 0.4257,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.6925207756232687,
|
| 1756 |
+
"grad_norm": 3.927248001098633,
|
| 1757 |
+
"learning_rate": 1.3485825461327812e-05,
|
| 1758 |
+
"loss": 0.4034,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6952908587257618,
|
| 1763 |
+
"grad_norm": 4.411158561706543,
|
| 1764 |
+
"learning_rate": 1.3263584182034849e-05,
|
| 1765 |
+
"loss": 0.4296,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6980609418282548,
|
| 1770 |
+
"grad_norm": 4.725271224975586,
|
| 1771 |
+
"learning_rate": 1.3042668494897804e-05,
|
| 1772 |
+
"loss": 0.4591,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7008310249307479,
|
| 1777 |
+
"grad_norm": 4.10283899307251,
|
| 1778 |
+
"learning_rate": 1.2823095897336969e-05,
|
| 1779 |
+
"loss": 0.4464,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.703601108033241,
|
| 1784 |
+
"grad_norm": 3.92057728767395,
|
| 1785 |
+
"learning_rate": 1.2604883780394499e-05,
|
| 1786 |
+
"loss": 0.431,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.7063711911357341,
|
| 1791 |
+
"grad_norm": 4.471419334411621,
|
| 1792 |
+
"learning_rate": 1.2388049427356943e-05,
|
| 1793 |
+
"loss": 0.4082,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.7091412742382271,
|
| 1798 |
+
"grad_norm": 4.321521282196045,
|
| 1799 |
+
"learning_rate": 1.2172610012386376e-05,
|
| 1800 |
+
"loss": 0.4472,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.7119113573407202,
|
| 1805 |
+
"grad_norm": 4.131201267242432,
|
| 1806 |
+
"learning_rate": 1.1958582599160073e-05,
|
| 1807 |
+
"loss": 0.427,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.7146814404432132,
|
| 1812 |
+
"grad_norm": 3.820596933364868,
|
| 1813 |
+
"learning_rate": 1.1745984139519071e-05,
|
| 1814 |
+
"loss": 0.4138,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.7174515235457064,
|
| 1819 |
+
"grad_norm": 4.444502353668213,
|
| 1820 |
+
"learning_rate": 1.153483147212544e-05,
|
| 1821 |
+
"loss": 0.4109,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.7202216066481995,
|
| 1826 |
+
"grad_norm": 4.224623203277588,
|
| 1827 |
+
"learning_rate": 1.1325141321128649e-05,
|
| 1828 |
+
"loss": 0.4202,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.7229916897506925,
|
| 1833 |
+
"grad_norm": 3.9233286380767822,
|
| 1834 |
+
"learning_rate": 1.11169302948409e-05,
|
| 1835 |
+
"loss": 0.4377,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.7257617728531855,
|
| 1840 |
+
"grad_norm": 4.564990997314453,
|
| 1841 |
+
"learning_rate": 1.0910214884421735e-05,
|
| 1842 |
+
"loss": 0.4184,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7285318559556787,
|
| 1847 |
+
"grad_norm": 3.8863961696624756,
|
| 1848 |
+
"learning_rate": 1.0705011462571842e-05,
|
| 1849 |
+
"loss": 0.4326,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7313019390581718,
|
| 1854 |
+
"grad_norm": 3.9038331508636475,
|
| 1855 |
+
"learning_rate": 1.050133628223623e-05,
|
| 1856 |
+
"loss": 0.4407,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7340720221606648,
|
| 1861 |
+
"grad_norm": 4.037172317504883,
|
| 1862 |
+
"learning_rate": 1.0299205475317022e-05,
|
| 1863 |
+
"loss": 0.4184,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7368421052631579,
|
| 1868 |
+
"grad_norm": 4.162524700164795,
|
| 1869 |
+
"learning_rate": 1.0098635051395648e-05,
|
| 1870 |
+
"loss": 0.437,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.739612188365651,
|
| 1875 |
+
"grad_norm": 3.8530848026275635,
|
| 1876 |
+
"learning_rate": 9.899640896464874e-06,
|
| 1877 |
+
"loss": 0.4172,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7423822714681441,
|
| 1882 |
+
"grad_norm": 4.2646284103393555,
|
| 1883 |
+
"learning_rate": 9.70223877167055e-06,
|
| 1884 |
+
"loss": 0.4331,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.7451523545706371,
|
| 1889 |
+
"grad_norm": 4.318920135498047,
|
| 1890 |
+
"learning_rate": 9.5064443120633e-06,
|
| 1891 |
+
"loss": 0.4275,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7479224376731302,
|
| 1896 |
+
"grad_norm": 3.900696277618408,
|
| 1897 |
+
"learning_rate": 9.312273025360084e-06,
|
| 1898 |
+
"loss": 0.4575,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7506925207756233,
|
| 1903 |
+
"grad_norm": 4.387091159820557,
|
| 1904 |
+
"learning_rate": 9.119740290716017e-06,
|
| 1905 |
+
"loss": 0.4565,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7534626038781164,
|
| 1910 |
+
"grad_norm": 4.237148761749268,
|
| 1911 |
+
"learning_rate": 8.928861357506193e-06,
|
| 1912 |
+
"loss": 0.4242,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7562326869806094,
|
| 1917 |
+
"grad_norm": 4.315036773681641,
|
| 1918 |
+
"learning_rate": 8.739651344117926e-06,
|
| 1919 |
+
"loss": 0.4222,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7590027700831025,
|
| 1924 |
+
"grad_norm": 4.285457134246826,
|
| 1925 |
+
"learning_rate": 8.552125236753309e-06,
|
| 1926 |
+
"loss": 0.4275,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7617728531855956,
|
| 1931 |
+
"grad_norm": 3.956871747970581,
|
| 1932 |
+
"learning_rate": 8.366297888242205e-06,
|
| 1933 |
+
"loss": 0.4026,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.7645429362880887,
|
| 1938 |
+
"grad_norm": 4.065526008605957,
|
| 1939 |
+
"learning_rate": 8.18218401686591e-06,
|
| 1940 |
+
"loss": 0.4079,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.7673130193905817,
|
| 1945 |
+
"grad_norm": 3.7003588676452637,
|
| 1946 |
+
"learning_rate": 7.999798205191323e-06,
|
| 1947 |
+
"loss": 0.4166,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7700831024930748,
|
| 1952 |
+
"grad_norm": 4.1025800704956055,
|
| 1953 |
+
"learning_rate": 7.819154898916042e-06,
|
| 1954 |
+
"loss": 0.4222,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7728531855955678,
|
| 1959 |
+
"grad_norm": 4.2863593101501465,
|
| 1960 |
+
"learning_rate": 7.640268405724095e-06,
|
| 1961 |
+
"loss": 0.4072,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.775623268698061,
|
| 1966 |
+
"grad_norm": 3.834885358810425,
|
| 1967 |
+
"learning_rate": 7.463152894152826e-06,
|
| 1968 |
+
"loss": 0.461,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.778393351800554,
|
| 1973 |
+
"grad_norm": 3.8217790126800537,
|
| 1974 |
+
"learning_rate": 7.287822392470604e-06,
|
| 1975 |
+
"loss": 0.4323,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.7811634349030471,
|
| 1980 |
+
"grad_norm": 3.9972996711730957,
|
| 1981 |
+
"learning_rate": 7.114290787565773e-06,
|
| 1982 |
+
"loss": 0.4443,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7839335180055401,
|
| 1987 |
+
"grad_norm": 4.411800861358643,
|
| 1988 |
+
"learning_rate": 6.942571823846755e-06,
|
| 1989 |
+
"loss": 0.4323,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7867036011080333,
|
| 1994 |
+
"grad_norm": 4.5063300132751465,
|
| 1995 |
+
"learning_rate": 6.772679102153395e-06,
|
| 1996 |
+
"loss": 0.4228,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7894736842105263,
|
| 2001 |
+
"grad_norm": 4.130795955657959,
|
| 2002 |
+
"learning_rate": 6.6046260786797585e-06,
|
| 2003 |
+
"loss": 0.4197,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.7922437673130194,
|
| 2008 |
+
"grad_norm": 4.331823348999023,
|
| 2009 |
+
"learning_rate": 6.438426063908331e-06,
|
| 2010 |
+
"loss": 0.4269,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7950138504155124,
|
| 2015 |
+
"grad_norm": 3.887230396270752,
|
| 2016 |
+
"learning_rate": 6.2740922215558e-06,
|
| 2017 |
+
"loss": 0.4402,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7977839335180056,
|
| 2022 |
+
"grad_norm": 4.196553707122803,
|
| 2023 |
+
"learning_rate": 6.11163756753037e-06,
|
| 2024 |
+
"loss": 0.4417,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.8005540166204986,
|
| 2029 |
+
"grad_norm": 4.163293361663818,
|
| 2030 |
+
"learning_rate": 5.951074968900941e-06,
|
| 2031 |
+
"loss": 0.4216,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.8033240997229917,
|
| 2036 |
+
"grad_norm": 4.441506385803223,
|
| 2037 |
+
"learning_rate": 5.792417142877902e-06,
|
| 2038 |
+
"loss": 0.415,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8060941828254847,
|
| 2043 |
+
"grad_norm": 4.561277866363525,
|
| 2044 |
+
"learning_rate": 5.635676655805913e-06,
|
| 2045 |
+
"loss": 0.4089,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.8088642659279779,
|
| 2050 |
+
"grad_norm": 4.779412746429443,
|
| 2051 |
+
"learning_rate": 5.480865922168612e-06,
|
| 2052 |
+
"loss": 0.4076,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.8116343490304709,
|
| 2057 |
+
"grad_norm": 4.437448024749756,
|
| 2058 |
+
"learning_rate": 5.3279972036053005e-06,
|
| 2059 |
+
"loss": 0.3955,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.814404432132964,
|
| 2064 |
+
"grad_norm": 4.1041035652160645,
|
| 2065 |
+
"learning_rate": 5.177082607939821e-06,
|
| 2066 |
+
"loss": 0.4441,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.817174515235457,
|
| 2071 |
+
"grad_norm": 4.502457141876221,
|
| 2072 |
+
"learning_rate": 5.028134088221509e-06,
|
| 2073 |
+
"loss": 0.4097,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.8199445983379502,
|
| 2078 |
+
"grad_norm": 4.109027862548828,
|
| 2079 |
+
"learning_rate": 4.881163441778526e-06,
|
| 2080 |
+
"loss": 0.4292,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.8227146814404432,
|
| 2085 |
+
"grad_norm": 4.413381099700928,
|
| 2086 |
+
"learning_rate": 4.736182309283401e-06,
|
| 2087 |
+
"loss": 0.4361,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.8254847645429363,
|
| 2092 |
+
"grad_norm": 4.096319198608398,
|
| 2093 |
+
"learning_rate": 4.593202173831094e-06,
|
| 2094 |
+
"loss": 0.4059,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.8282548476454293,
|
| 2099 |
+
"grad_norm": 4.19406270980835,
|
| 2100 |
+
"learning_rate": 4.452234360029446e-06,
|
| 2101 |
+
"loss": 0.4164,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.8310249307479224,
|
| 2106 |
+
"grad_norm": 3.696166753768921,
|
| 2107 |
+
"learning_rate": 4.3132900331022695e-06,
|
| 2108 |
+
"loss": 0.4374,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.8337950138504155,
|
| 2113 |
+
"grad_norm": 4.229091167449951,
|
| 2114 |
+
"learning_rate": 4.176380198004969e-06,
|
| 2115 |
+
"loss": 0.4347,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.8365650969529086,
|
| 2120 |
+
"grad_norm": 4.072672367095947,
|
| 2121 |
+
"learning_rate": 4.041515698552953e-06,
|
| 2122 |
+
"loss": 0.4268,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.8393351800554016,
|
| 2127 |
+
"grad_norm": 4.1049981117248535,
|
| 2128 |
+
"learning_rate": 3.908707216562705e-06,
|
| 2129 |
+
"loss": 0.4186,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.8421052631578947,
|
| 2134 |
+
"grad_norm": 4.255549907684326,
|
| 2135 |
+
"learning_rate": 3.7779652710057797e-06,
|
| 2136 |
+
"loss": 0.4232,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.8448753462603878,
|
| 2141 |
+
"grad_norm": 4.021961688995361,
|
| 2142 |
+
"learning_rate": 3.649300217175661e-06,
|
| 2143 |
+
"loss": 0.408,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.8476454293628809,
|
| 2148 |
+
"grad_norm": 4.191747665405273,
|
| 2149 |
+
"learning_rate": 3.522722245867548e-06,
|
| 2150 |
+
"loss": 0.4251,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.850415512465374,
|
| 2155 |
+
"grad_norm": 4.252630233764648,
|
| 2156 |
+
"learning_rate": 3.398241382571237e-06,
|
| 2157 |
+
"loss": 0.395,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.853185595567867,
|
| 2162 |
+
"grad_norm": 3.839334726333618,
|
| 2163 |
+
"learning_rate": 3.2758674866770278e-06,
|
| 2164 |
+
"loss": 0.4218,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.8559556786703602,
|
| 2169 |
+
"grad_norm": 4.2264838218688965,
|
| 2170 |
+
"learning_rate": 3.155610250694866e-06,
|
| 2171 |
+
"loss": 0.4468,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.8587257617728532,
|
| 2176 |
+
"grad_norm": 3.8925042152404785,
|
| 2177 |
+
"learning_rate": 3.03747919948659e-06,
|
| 2178 |
+
"loss": 0.4315,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8614958448753463,
|
| 2183 |
+
"grad_norm": 4.53446102142334,
|
| 2184 |
+
"learning_rate": 2.9214836895115947e-06,
|
| 2185 |
+
"loss": 0.4353,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8642659279778393,
|
| 2190 |
+
"grad_norm": 4.308258533477783,
|
| 2191 |
+
"learning_rate": 2.8076329080857277e-06,
|
| 2192 |
+
"loss": 0.4441,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8670360110803325,
|
| 2197 |
+
"grad_norm": 4.659024715423584,
|
| 2198 |
+
"learning_rate": 2.6959358726535964e-06,
|
| 2199 |
+
"loss": 0.4165,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8698060941828255,
|
| 2204 |
+
"grad_norm": 4.282739639282227,
|
| 2205 |
+
"learning_rate": 2.5864014300743964e-06,
|
| 2206 |
+
"loss": 0.3965,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.8725761772853186,
|
| 2211 |
+
"grad_norm": 4.223807334899902,
|
| 2212 |
+
"learning_rate": 2.479038255921171e-06,
|
| 2213 |
+
"loss": 0.405,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8753462603878116,
|
| 2218 |
+
"grad_norm": 4.042827129364014,
|
| 2219 |
+
"learning_rate": 2.373854853793671e-06,
|
| 2220 |
+
"loss": 0.4223,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8781163434903048,
|
| 2225 |
+
"grad_norm": 3.7920212745666504,
|
| 2226 |
+
"learning_rate": 2.2708595546448442e-06,
|
| 2227 |
+
"loss": 0.4104,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8808864265927978,
|
| 2232 |
+
"grad_norm": 4.02117395401001,
|
| 2233 |
+
"learning_rate": 2.170060516121012e-06,
|
| 2234 |
+
"loss": 0.401,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8836565096952909,
|
| 2239 |
+
"grad_norm": 4.442043304443359,
|
| 2240 |
+
"learning_rate": 2.071465721915694e-06,
|
| 2241 |
+
"loss": 0.4345,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8864265927977839,
|
| 2246 |
+
"grad_norm": 4.755099773406982,
|
| 2247 |
+
"learning_rate": 1.9750829811373307e-06,
|
| 2248 |
+
"loss": 0.3976,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.889196675900277,
|
| 2253 |
+
"grad_norm": 4.009551525115967,
|
| 2254 |
+
"learning_rate": 1.8809199276907196e-06,
|
| 2255 |
+
"loss": 0.4175,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.8919667590027701,
|
| 2260 |
+
"grad_norm": 3.991119623184204,
|
| 2261 |
+
"learning_rate": 1.7889840196724083e-06,
|
| 2262 |
+
"loss": 0.4082,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.8947368421052632,
|
| 2267 |
+
"grad_norm": 3.973052740097046,
|
| 2268 |
+
"learning_rate": 1.6992825387799816e-06,
|
| 2269 |
+
"loss": 0.4091,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.8975069252077562,
|
| 2274 |
+
"grad_norm": 4.348862171173096,
|
| 2275 |
+
"learning_rate": 1.6118225897352967e-06,
|
| 2276 |
+
"loss": 0.4279,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.9002770083102493,
|
| 2281 |
+
"grad_norm": 4.407948017120361,
|
| 2282 |
+
"learning_rate": 1.5266110997218009e-06,
|
| 2283 |
+
"loss": 0.4195,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.9030470914127424,
|
| 2288 |
+
"grad_norm": 4.2243428230285645,
|
| 2289 |
+
"learning_rate": 1.4436548178358188e-06,
|
| 2290 |
+
"loss": 0.4193,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.9058171745152355,
|
| 2295 |
+
"grad_norm": 4.009667873382568,
|
| 2296 |
+
"learning_rate": 1.3629603145520554e-06,
|
| 2297 |
+
"loss": 0.3961,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.9085872576177285,
|
| 2302 |
+
"grad_norm": 4.0680060386657715,
|
| 2303 |
+
"learning_rate": 1.2845339812031397e-06,
|
| 2304 |
+
"loss": 0.4235,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.9113573407202216,
|
| 2309 |
+
"grad_norm": 3.899540901184082,
|
| 2310 |
+
"learning_rate": 1.208382029473426e-06,
|
| 2311 |
+
"loss": 0.4234,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.9141274238227147,
|
| 2316 |
+
"grad_norm": 4.105435848236084,
|
| 2317 |
+
"learning_rate": 1.1345104909069947e-06,
|
| 2318 |
+
"loss": 0.4522,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.9168975069252078,
|
| 2323 |
+
"grad_norm": 4.221428871154785,
|
| 2324 |
+
"learning_rate": 1.062925216429953e-06,
|
| 2325 |
+
"loss": 0.4117,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.9196675900277008,
|
| 2330 |
+
"grad_norm": 4.16056489944458,
|
| 2331 |
+
"learning_rate": 9.936318758869877e-07,
|
| 2332 |
+
"loss": 0.4211,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.9224376731301939,
|
| 2337 |
+
"grad_norm": 4.145576000213623,
|
| 2338 |
+
"learning_rate": 9.26635957592299e-07,
|
| 2339 |
+
"loss": 0.4086,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.925207756232687,
|
| 2344 |
+
"grad_norm": 4.158425807952881,
|
| 2345 |
+
"learning_rate": 8.619427678949165e-07,
|
| 2346 |
+
"loss": 0.4123,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.9279778393351801,
|
| 2351 |
+
"grad_norm": 3.9155080318450928,
|
| 2352 |
+
"learning_rate": 7.995574307583975e-07,
|
| 2353 |
+
"loss": 0.4271,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.9307479224376731,
|
| 2358 |
+
"grad_norm": 4.067850589752197,
|
| 2359 |
+
"learning_rate": 7.394848873550064e-07,
|
| 2360 |
+
"loss": 0.4156,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.9335180055401662,
|
| 2365 |
+
"grad_norm": 4.133598804473877,
|
| 2366 |
+
"learning_rate": 6.817298956743356e-07,
|
| 2367 |
+
"loss": 0.4449,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.9362880886426593,
|
| 2372 |
+
"grad_norm": 4.627798080444336,
|
| 2373 |
+
"learning_rate": 6.262970301464632e-07,
|
| 2374 |
+
"loss": 0.4143,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.9390581717451524,
|
| 2379 |
+
"grad_norm": 4.110208034515381,
|
| 2380 |
+
"learning_rate": 5.731906812796395e-07,
|
| 2381 |
+
"loss": 0.4238,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.9418282548476454,
|
| 2386 |
+
"grad_norm": 4.190328598022461,
|
| 2387 |
+
"learning_rate": 5.22415055312545e-07,
|
| 2388 |
+
"loss": 0.3903,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.9445983379501385,
|
| 2393 |
+
"grad_norm": 4.284666538238525,
|
| 2394 |
+
"learning_rate": 4.7397417388111766e-07,
|
| 2395 |
+
"loss": 0.3984,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.9473684210526315,
|
| 2400 |
+
"grad_norm": 4.148164749145508,
|
| 2401 |
+
"learning_rate": 4.278718737000564e-07,
|
| 2402 |
+
"loss": 0.4286,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.9501385041551247,
|
| 2407 |
+
"grad_norm": 4.629441261291504,
|
| 2408 |
+
"learning_rate": 3.8411180625890886e-07,
|
| 2409 |
+
"loss": 0.409,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.9529085872576177,
|
| 2414 |
+
"grad_norm": 4.130425453186035,
|
| 2415 |
+
"learning_rate": 3.4269743753287175e-07,
|
| 2416 |
+
"loss": 0.423,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.9556786703601108,
|
| 2421 |
+
"grad_norm": 3.745643377304077,
|
| 2422 |
+
"learning_rate": 3.03632047708281e-07,
|
| 2423 |
+
"loss": 0.4262,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.9584487534626038,
|
| 2428 |
+
"grad_norm": 4.220255374908447,
|
| 2429 |
+
"learning_rate": 2.669187309227794e-07,
|
| 2430 |
+
"loss": 0.4183,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.961218836565097,
|
| 2435 |
+
"grad_norm": 3.9583799839019775,
|
| 2436 |
+
"learning_rate": 2.3256039502027592e-07,
|
| 2437 |
+
"loss": 0.4221,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.96398891966759,
|
| 2442 |
+
"grad_norm": 3.985778331756592,
|
| 2443 |
+
"learning_rate": 2.005597613206167e-07,
|
| 2444 |
+
"loss": 0.4088,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.9667590027700831,
|
| 2449 |
+
"grad_norm": 4.007012844085693,
|
| 2450 |
+
"learning_rate": 1.7091936440405742e-07,
|
| 2451 |
+
"loss": 0.4379,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9695290858725761,
|
| 2456 |
+
"grad_norm": 4.563694000244141,
|
| 2457 |
+
"learning_rate": 1.436415519104972e-07,
|
| 2458 |
+
"loss": 0.4149,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9722991689750693,
|
| 2463 |
+
"grad_norm": 4.10459041595459,
|
| 2464 |
+
"learning_rate": 1.1872848435356076e-07,
|
| 2465 |
+
"loss": 0.4297,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9750692520775623,
|
| 2470 |
+
"grad_norm": 4.2543487548828125,
|
| 2471 |
+
"learning_rate": 9.618213494945183e-08,
|
| 2472 |
+
"loss": 0.3925,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9778393351800554,
|
| 2477 |
+
"grad_norm": 4.755935192108154,
|
| 2478 |
+
"learning_rate": 7.600428946067828e-08,
|
| 2479 |
+
"loss": 0.4242,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.9806094182825484,
|
| 2484 |
+
"grad_norm": 4.033304214477539,
|
| 2485 |
+
"learning_rate": 5.819654605461522e-08,
|
| 2486 |
+
"loss": 0.4195,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.9833795013850416,
|
| 2491 |
+
"grad_norm": 3.9951982498168945,
|
| 2492 |
+
"learning_rate": 4.276031517691292e-08,
|
| 2493 |
+
"loss": 0.4187,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.9861495844875346,
|
| 2498 |
+
"grad_norm": 4.105679988861084,
|
| 2499 |
+
"learning_rate": 2.9696819439802846e-08,
|
| 2500 |
+
"loss": 0.4328,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.9889196675900277,
|
| 2505 |
+
"grad_norm": 4.221299648284912,
|
| 2506 |
+
"learning_rate": 1.900709352524177e-08,
|
| 2507 |
+
"loss": 0.4281,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9916897506925207,
|
| 2512 |
+
"grad_norm": 4.06874418258667,
|
| 2513 |
+
"learning_rate": 1.0691984102964014e-08,
|
| 2514 |
+
"loss": 0.4126,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9944598337950139,
|
| 2519 |
+
"grad_norm": 4.161386489868164,
|
| 2520 |
+
"learning_rate": 4.7521497634317454e-09,
|
| 2521 |
+
"loss": 0.39,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.997229916897507,
|
| 2526 |
+
"grad_norm": 4.167057991027832,
|
| 2527 |
+
"learning_rate": 1.1880609656733653e-09,
|
| 2528 |
+
"loss": 0.4371,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 1.0,
|
| 2533 |
+
"grad_norm": 3.841407537460327,
|
| 2534 |
+
"learning_rate": 0.0,
|
| 2535 |
+
"loss": 0.4198,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
}
|
| 2538 |
+
],
|
| 2539 |
+
"logging_steps": 1,
|
| 2540 |
+
"max_steps": 361,
|
| 2541 |
+
"num_input_tokens_seen": 0,
|
| 2542 |
+
"num_train_epochs": 1,
|
| 2543 |
+
"save_steps": 100,
|
| 2544 |
+
"stateful_callbacks": {
|
| 2545 |
+
"TrainerControl": {
|
| 2546 |
+
"args": {
|
| 2547 |
+
"should_epoch_stop": false,
|
| 2548 |
+
"should_evaluate": false,
|
| 2549 |
+
"should_log": false,
|
| 2550 |
+
"should_save": true,
|
| 2551 |
+
"should_training_stop": true
|
| 2552 |
+
},
|
| 2553 |
+
"attributes": {}
|
| 2554 |
+
}
|
| 2555 |
+
},
|
| 2556 |
+
"total_flos": 8.230029212895805e+17,
|
| 2557 |
+
"train_batch_size": 9,
|
| 2558 |
+
"trial_name": null,
|
| 2559 |
+
"trial_params": null
|
| 2560 |
+
}
|
checkpoint-361/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab1cdf662fe04dee407e707c51291816e5f02af9c61f64c59aab73962edcc8a6
|
| 3 |
+
size 5560
|