metadata
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- unsloth
- generated_from_trainer
base_model: unsloth/llama-2-7b-chat-bnb-4bit
model-index:
- name: llama2-chat_10000_200
results: []
llama2-chat_10000_200
This model is a fine-tuned version of unsloth/llama-2-7b-chat-bnb-4bit on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0901
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: 8
- eval_batch_size: 4
- seed: 3407
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 5
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.379 | 0.31 | 48 | 1.0400 |
| 1.059 | 0.61 | 96 | 1.0091 |
| 1.0438 | 0.92 | 144 | 0.9990 |
| 0.9934 | 1.23 | 192 | 0.9968 |
| 0.9749 | 1.54 | 240 | 0.9926 |
| 0.9778 | 1.84 | 288 | 0.9862 |
| 0.9443 | 2.15 | 336 | 1.0046 |
| 0.8913 | 2.46 | 384 | 1.0017 |
| 0.8908 | 2.76 | 432 | 0.9996 |
| 0.8708 | 3.07 | 480 | 1.0339 |
| 0.7958 | 3.38 | 528 | 1.0386 |
| 0.8025 | 3.69 | 576 | 1.0386 |
| 0.8099 | 3.99 | 624 | 1.0386 |
| 0.7191 | 4.3 | 672 | 1.0913 |
| 0.7138 | 4.61 | 720 | 1.0926 |
| 0.723 | 4.92 | 768 | 1.0901 |
Framework versions
- PEFT 0.10.0
- Transformers 4.39.3
- Pytorch 2.2.2+cu121
- Datasets 2.16.0
- Tokenizers 0.15.2