WeniGPT-2.1.1-zephyr-7b-beta-BitsandBytes-LLM-Base-1.0.1-6k_evol_complexity_increase_steps

This model is a fine-tuned version of HuggingFaceH4/zephyr-7b-beta on the generator dataset. It achieves the following results on the evaluation set:

• Loss: 0.3968

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: 16
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.03
• training_steps: 252
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.99	34	0.4474
0.563	2.0	69	0.4172
0.4141	2.99	103	0.4064
0.4141	4.0	138	0.4008
0.3912	4.99	172	0.3978
0.3778	6.0	207	0.3971
0.3778	6.99	241	0.3968
0.3722	7.3	252	0.3968

Framework versions

• PEFT 0.7.1
• Transformers 4.37.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.16.1
• Tokenizers 0.15.0