codellama-hugcoder

This model is a fine-tuned version of openai-community/gpt2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.2146

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss
22.1932	0.1	100	3.1842
22.9842	0.2	200	1.7973
20.6306	0.3	300	1.6439
16.7058	0.4	400	1.4211
17.5604	0.5	500	1.3250
17.0438	0.6	600	1.2830
14.083	0.7	700	1.2410
13.923	0.8	800	1.2257
15.3756	0.9	900	1.2145
13.542	1.0	1000	1.2146

Framework versions

• PEFT 0.13.3.dev0
• Transformers 4.46.0.dev0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1