Hugging Face's logo

oldiday
/
a30680df-bea9-4787-af5d-52a195cad69d

PEFT
Safetensors
mixtral
axolotl
Generated from Trainer
Model card Files Community
a30680df-bea9-4787-af5d-52a195cad69d / README.md
oldiday's picture
oldiday
End of training
23d7868 verified
preview code
|
raw
history blame contribute delete
4.27 kB
metadata
library_name: peft
base_model: TitanML/tiny-mixtral
tags:
  - axolotl
  - generated_from_trainer
model-index:
  - name: a30680df-bea9-4787-af5d-52a195cad69d
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: TitanML/tiny-mixtral
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 27baa71b1068ef49_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/27baa71b1068ef49_train_data.json
  type:
    field_input: my_solu
    field_instruction: prompt
    field_output: solution
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 4
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
gradient_clipping: 1.0
group_by_length: false
hub_model_id: oldiday/a30680df-bea9-4787-af5d-52a195cad69d
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 5.0e-05
load_in_4bit: false
load_in_8bit: false
local_rank: 0
logging_steps: 3
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine
max_steps: 100
micro_batch_size: 8
mlflow_experiment_name: /tmp/27baa71b1068ef49_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 4
sequence_len: 1024
special_tokens:
  pad_token: </s>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: techspear-hub
wandb_mode: online
wandb_name: d527ff26-28a8-406b-842c-388110bbc2e8
wandb_project: Gradients-On-Six
wandb_run: your_name
wandb_runid: d527ff26-28a8-406b-842c-388110bbc2e8
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

a30680df-bea9-4787-af5d-52a195cad69d

This model is a fine-tuned version of TitanML/tiny-mixtral on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 8.8788

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • 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: 10
  • training_steps: 100

Training results

Training Loss Epoch Step Validation Loss
No log 0.0001 1 10.5942
10.5738 0.0009 9 10.5639
10.4435 0.0017 18 10.3698
10.1468 0.0026 27 10.0672
9.8311 0.0034 36 9.7385
9.4899 0.0043 45 9.4465
9.2884 0.0051 54 9.2170
9.1081 0.0060 63 9.0550
8.957 0.0068 72 8.9552
8.9262 0.0077 81 8.9030
8.9591 0.0085 90 8.8825
8.8878 0.0094 99 8.8788

Framework versions

  • PEFT 0.13.2
  • Transformers 4.46.0
  • Pytorch 2.5.0+cu124
  • Datasets 3.0.1
  • Tokenizers 0.20.1