mistral_texmin_generation

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0188

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: True
• bnb_4bit_compute_dtype: bfloat16

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_steps: 0.03
• training_steps: 100

Training results

Training Loss	Epoch	Step	Validation Loss
0.4867	1.05	20	0.2200
0.0692	2.11	40	0.0339
0.0153	3.16	60	0.0216
0.0082	4.21	80	0.0200
0.0066	5.26	100	0.0188

Framework versions

• PEFT 0.7.0
• Transformers 4.36.0
• Pytorch 2.1.1+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0