|
|
--- |
|
|
library_name: peft |
|
|
license: llama3.1 |
|
|
base_model: mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated |
|
|
tags: |
|
|
- axolotl |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- dset_comp3.0_sortpatent_count_pat400_in5_5000.jsonl |
|
|
model-index: |
|
|
- name: alpha32_r64_lr0.00002_Meta-Llama-3.1-_dset_comp3.0_sortpatent_count_pat400_in5_5000 |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
|
|
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) |
|
|
<details><summary>See axolotl config</summary> |
|
|
|
|
|
axolotl version: `0.9.1` |
|
|
```yaml |
|
|
base_model: mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated |
|
|
load_in_8bit: false |
|
|
load_in_4bit: true |
|
|
adapter: qlora |
|
|
wandb_name: Meta-Llama-3.1-_outputs_axolotl_ft_alpha32_r64_lr0.00002_Meta-Llama-3.1-_dset_comp3.0_sortpatent_count_pat400_in5_5000 |
|
|
output_dir: ./outputs/out/Meta-Llama-3.1-_outputs_axolotl_ft_alpha32_r64_lr0.00002_Meta-Llama-3.1-_dset_comp3.0_sortpatent_count_pat400_in5_5000 |
|
|
hub_model_id: cgifbribcgfbi/alpha32_r64_lr0.00002_Meta-Llama-3.1-_dset_comp3.0_sortpatent_count_pat400_in5_5000 |
|
|
|
|
|
tokenizer_type: AutoTokenizer |
|
|
push_dataset_to_hub: |
|
|
strict: false |
|
|
|
|
|
datasets: |
|
|
- path: dset_comp3.0_sortpatent_count_pat400_in5_5000.jsonl |
|
|
type: chat_template |
|
|
field_messages: messages |
|
|
|
|
|
dataset_prepared_path: last_run_prepared |
|
|
val_set_size: 0.04 |
|
|
save_safetensors: true |
|
|
|
|
|
sequence_len: 2700 |
|
|
sample_packing: true |
|
|
pad_to_sequence_len: true |
|
|
|
|
|
lora_r: 64 |
|
|
lora_alpha: 32 |
|
|
lora_dropout: 0.05 |
|
|
lora_target_modules: |
|
|
lora_target_linear: true |
|
|
|
|
|
wandb_mode: |
|
|
wandb_project: finetune-sweep |
|
|
wandb_entity: gpoisjgqetpadsfke |
|
|
wandb_watch: |
|
|
wandb_run_id: |
|
|
wandb_log_model: |
|
|
|
|
|
gradient_accumulation_steps: 1 |
|
|
micro_batch_size: 4 # This will be automatically adjusted based on available GPU memory |
|
|
num_epochs: 4 |
|
|
optimizer: adamw_torch_fused |
|
|
lr_scheduler: cosine |
|
|
learning_rate: 0.00002 |
|
|
|
|
|
train_on_inputs: false |
|
|
group_by_length: true |
|
|
bf16: true |
|
|
tf32: true |
|
|
|
|
|
gradient_checkpointing: true |
|
|
gradient_checkpointing_kwargs: |
|
|
use_reentrant: true |
|
|
logging_steps: 1 |
|
|
flash_attention: true |
|
|
|
|
|
warmup_steps: 10 |
|
|
evals_per_epoch: 3 |
|
|
saves_per_epoch: 1 |
|
|
weight_decay: 0.01 |
|
|
fsdp: |
|
|
- full_shard |
|
|
- auto_wrap |
|
|
fsdp_config: |
|
|
fsdp_limit_all_gathers: true |
|
|
fsdp_sync_module_states: true |
|
|
fsdp_offload_params: false |
|
|
fsdp_use_orig_params: false |
|
|
fsdp_cpu_ram_efficient_loading: true |
|
|
fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP |
|
|
fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer |
|
|
fsdp_state_dict_type: FULL_STATE_DICT |
|
|
fsdp_sharding_strategy: FULL_SHARD |
|
|
special_tokens: |
|
|
pad_token: <|finetune_right_pad_id|> |
|
|
``` |
|
|
|
|
|
</details><br> |
|
|
|
|
|
# alpha32_r64_lr0.00002_Meta-Llama-3.1-_dset_comp3.0_sortpatent_count_pat400_in5_5000 |
|
|
|
|
|
This model is a fine-tuned version of [mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated](https://huggingface.co/mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated) on the dset_comp3.0_sortpatent_count_pat400_in5_5000.jsonl dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.4583 |
|
|
|
|
|
## 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: 2e-05 |
|
|
- train_batch_size: 4 |
|
|
- eval_batch_size: 4 |
|
|
- seed: 42 |
|
|
- distributed_type: multi-GPU |
|
|
- num_devices: 4 |
|
|
- total_train_batch_size: 16 |
|
|
- total_eval_batch_size: 16 |
|
|
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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 |
|
|
- num_epochs: 4.0 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:------:|:----:|:---------------:| |
|
|
| 0.7 | 0.0061 | 1 | 0.8766 | |
|
|
| 0.6414 | 0.3354 | 55 | 0.6293 | |
|
|
| 0.5608 | 0.6707 | 110 | 0.5473 | |
|
|
| 0.4733 | 1.0061 | 165 | 0.5161 | |
|
|
| 0.5142 | 1.3415 | 220 | 0.4954 | |
|
|
| 0.4771 | 1.6768 | 275 | 0.4824 | |
|
|
| 0.423 | 2.0122 | 330 | 0.4750 | |
|
|
| 0.4375 | 2.3476 | 385 | 0.4676 | |
|
|
| 0.4311 | 2.6829 | 440 | 0.4630 | |
|
|
| 0.4019 | 3.0183 | 495 | 0.4620 | |
|
|
| 0.4726 | 3.3537 | 550 | 0.4589 | |
|
|
| 0.4677 | 3.6890 | 605 | 0.4583 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.15.2 |
|
|
- Transformers 4.51.3 |
|
|
- Pytorch 2.6.0+cu124 |
|
|
- Datasets 3.5.1 |
|
|
- Tokenizers 0.21.1 |
