metadata
base_model: NousResearch/Llama-2-7b-chat-hf
tags:
- generated_from_trainer
model-index:
- name: Cosine_matric_llama2_prompt1
results: []
library_name: peft
Cosine_matric_llama2_prompt1
This model is a fine-tuned version of NousResearch/Llama-2-7b-chat-hf on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9479
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
The following bitsandbytes quantization config was used during training:
- quant_method: bitsandbytes
- _load_in_8bit: False
- _load_in_4bit: True
- llm_int8_threshold: 6.0
- llm_int8_skip_modules: None
- llm_int8_enable_fp32_cpu_offload: False
- llm_int8_has_fp16_weight: False
- bnb_4bit_quant_type: nf4
- bnb_4bit_use_double_quant: False
- bnb_4bit_compute_dtype: float16
- load_in_4bit: True
- load_in_8bit: False
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.0276 | 0.4 | 384 | 0.9479 |
| 1.0277 | 0.8 | 768 | 0.9479 |
| 1.0213 | 1.2 | 1152 | 0.9479 |
| 1.0277 | 1.6 | 1536 | 0.9479 |
| 0.8449 | 2.0 | 1920 | 0.9479 |
Framework versions
- PEFT 0.4.0
- Transformers 4.38.2
- Pytorch 2.1.2
- Datasets 2.1.0
- Tokenizers 0.15.2