End of training

README.md

@@ -66,7 +66,7 @@ lora_model_dir: null
 lora_r: 8
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 1
+max_steps: 50
 micro_batch_size: 2
 mlflow_experiment_name: /tmp/2cb9619ed101b972_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 8b2b0e39-7daa-495b-889a-e527ce88f772
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 8b2b0e39-7daa-495b-889a-e527ce88f772
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # d0845773-3993-445e-b789-0a1a90528ca7
 
 This model is a fine-tuned version of [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2971
 
 ## Model description
 
@@ -128,14 +130,19 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 8
 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
-- lr_scheduler_warmup_steps: 2
-- training_steps: 1
+- lr_scheduler_warmup_steps: 10
+- training_steps: 50
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0001 | 1    | 4.4844          |
+| 3.1762        | 0.0011 | 10   | 2.7510          |
+| 1.6153        | 0.0022 | 20   | 1.4343          |
+| 0.5395        | 0.0032 | 30   | 0.6099          |
+| 0.3597        | 0.0043 | 40   | 0.3144          |
+| 0.3599        | 0.0054 | 50   | 0.2971          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15b5ccecde0674d8348676835c3b47631e475886c7d93c09ec21fc1e2d1f8049
+oid sha256:ae6775a48809d22985a92fb476ffe981040605725f7ac955f21049262d22ee0b
 size 60048618