|
--- |
|
license: apache-2.0 |
|
library_name: peft |
|
tags: |
|
- generated_from_trainer |
|
base_model: openlm-research/open_llama_3b_v2 |
|
model-index: |
|
- name: qlora-out |
|
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/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl) |
|
<details><summary>See axolotl config</summary> |
|
|
|
axolotl version: `0.4.0` |
|
```yaml |
|
base_model: openlm-research/open_llama_3b_v2 |
|
model_type: LlamaForCausalLM |
|
tokenizer_type: LlamaTokenizer |
|
load_in_8bit: false |
|
load_in_4bit: true |
|
strict: false |
|
push_dataset_to_hub: |
|
datasets: |
|
- path: teknium/GPT4-LLM-Cleaned |
|
type: alpaca |
|
dataset_prepared_path: |
|
val_set_size: 0.05 |
|
adapter: qlora |
|
lora_model_dir: |
|
sequence_len: 1024 |
|
sample_packing: true |
|
lora_r: 8 |
|
lora_alpha: 32 |
|
lora_dropout: 0.05 |
|
lora_target_modules: |
|
lora_target_linear: true |
|
lora_fan_in_fan_out: |
|
wandb_project: |
|
wandb_entity: |
|
wandb_watch: |
|
wandb_name: |
|
wandb_log_model: |
|
output_dir: ./qlora-out |
|
gradient_accumulation_steps: 1 |
|
micro_batch_size: 2 |
|
num_epochs: 4 |
|
optimizer: paged_adamw_32bit |
|
torchdistx_path: |
|
lr_scheduler: cosine |
|
learning_rate: 0.0002 |
|
train_on_inputs: false |
|
group_by_length: false |
|
bf16: false |
|
fp16: true |
|
tf32: false |
|
gradient_checkpointing: true |
|
early_stopping_patience: |
|
resume_from_checkpoint: |
|
local_rank: |
|
logging_steps: 1 |
|
xformers_attention: |
|
flash_attention: true |
|
gptq_groupsize: |
|
gptq_model_v1: |
|
warmup_steps: 20 |
|
evals_per_epoch: 4 |
|
saves_per_epoch: 1 |
|
debug: |
|
deepspeed: |
|
weight_decay: 0.1 |
|
fsdp: |
|
fsdp_config: |
|
special_tokens: |
|
bos_token: "<s>" |
|
eos_token: "</s>" |
|
unk_token: "<unk>" |
|
|
|
``` |
|
|
|
</details><br> |
|
|
|
# qlora-out |
|
|
|
This model is a fine-tuned version of [openlm-research/open_llama_3b_v2](https://huggingface.co/openlm-research/open_llama_3b_v2) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.0305 |
|
|
|
## 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: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 2 |
|
- total_train_batch_size: 4 |
|
- total_eval_batch_size: 4 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 20 |
|
- num_epochs: 4 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.3161 | 0.0 | 1 | 1.5593 | |
|
| 1.1396 | 0.25 | 655 | 1.0154 | |
|
| 1.107 | 0.5 | 1310 | 0.9923 | |
|
| 1.1086 | 0.75 | 1965 | 0.9727 | |
|
| 0.9957 | 1.0 | 2620 | 0.9599 | |
|
| 1.0171 | 1.23 | 3275 | 0.9603 | |
|
| 0.7529 | 1.48 | 3930 | 0.9561 | |
|
| 1.1053 | 1.73 | 4585 | 0.9523 | |
|
| 0.8667 | 1.98 | 5240 | 0.9470 | |
|
| 0.8547 | 2.21 | 5895 | 0.9852 | |
|
| 0.8283 | 2.46 | 6550 | 0.9837 | |
|
| 1.0088 | 2.71 | 7205 | 0.9850 | |
|
| 0.8609 | 2.96 | 7860 | 0.9807 | |
|
| 0.8617 | 3.19 | 8515 | 1.0244 | |
|
| 0.4939 | 3.44 | 9170 | 1.0304 | |
|
| 0.6502 | 3.7 | 9825 | 1.0305 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.10.0 |
|
- Transformers 4.40.0.dev0 |
|
- Pytorch 2.1.2+cu118 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.0 |
