falcon-7b-sharded-bf16-finetuned-html-code-generation

This model is a fine-tuned version of ybelkada/falcon-7b-sharded-bf16 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.2896

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: 2
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.03
• training_steps: 320

Training results

Training Loss	Epoch	Step	Validation Loss
2.0689	0.1794	20	1.7177
1.754	0.3587	40	1.3882
1.6143	0.5381	60	1.3242
1.7434	0.7175	80	1.2945
1.3572	0.8969	100	1.2594
1.386	1.0762	120	1.2262
1.3228	1.2556	140	1.3785
1.2054	1.4350	160	1.2069
1.1205	1.6143	180	1.2393
1.1729	1.7937	200	1.1976
1.3121	1.9731	220	1.2099
0.912	2.1525	240	1.2399
0.7858	2.3318	260	1.2896

Framework versions

• PEFT 0.10.1.dev0
• Transformers 4.40.0
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1