Hugging Face's logo

masatochi
/
tuning-364a1e79-e5ec-4e64-ad45-fd532a9c377e

PEFT
Safetensors
llama
axolotl
Generated from Trainer
8-bit precision
bitsandbytes
Model card Files Community
tuning-364a1e79-e5ec-4e64-ad45-fd532a9c377e / README.md
masatochi's picture
masatochi
End of training
1396aa6 verified
preview code
|
raw
history blame contribute delete
3.75 kB
metadata
library_name: peft
license: llama3.1
base_model: shenzhi-wang/Llama3.1-8B-Chinese-Chat
tags:
  - axolotl
  - generated_from_trainer
model-index:
  - name: tuning-364a1e79-e5ec-4e64-ad45-fd532a9c377e
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: shenzhi-wang/Llama3.1-8B-Chinese-Chat
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - alpaca-cleaned_train_data.json
  ds_type: json
  path: /workspace/input_data/alpaca-cleaned_train_data.json
  type:
    field_input: input
    field_instruction: instruction
    field_output: output
    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: 2
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: masatochi/tuning-364a1e79-e5ec-4e64-ad45-fd532a9c377e
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: true
local_rank: null
logging_steps: 1
lora_alpha: 16
lora_dropout: 0.06
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 8
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 3
mlflow_experiment_name: /tmp/alpaca-cleaned_train_data.json
model_type: LlamaForCausalLM
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
save_steps: 5
save_strategy: steps
sequence_len: 4096
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
val_set_size: 0.05
wandb_entity: lkotbimehdi
wandb_mode: online
wandb_project: lko
wandb_run: miner_id_24
wandb_runid: 364a1e79-e5ec-4e64-ad45-fd532a9c377e
warmup_steps: 30
weight_decay: 0.0
xformers_attention: null

tuning-364a1e79-e5ec-4e64-ad45-fd532a9c377e

This model is a fine-tuned version of shenzhi-wang/Llama3.1-8B-Chinese-Chat on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.9933

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: 0.0002
  • train_batch_size: 3
  • eval_batch_size: 3
  • seed: 42
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 24
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 30
  • training_steps: 200

Training results

Training Loss Epoch Step Validation Loss
1.2369 0.0005 1 1.3036
1.1428 0.0166 34 1.0232
1.0508 0.0333 68 1.0058
0.9603 0.0499 102 0.9993
1.0164 0.0665 136 0.9951
0.843 0.0831 170 0.9933

Framework versions

  • PEFT 0.13.2
  • Transformers 4.45.2
  • Pytorch 2.4.1+cu124
  • Datasets 3.0.1
  • Tokenizers 0.20.1