f4b93473-10a3-4baf-bbb9-5db9c05eda6e

This model is a fine-tuned version of EleutherAI/pythia-410m-deduped on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.9434

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: 0.000201
• train_batch_size: 4
• eval_batch_size: 4
• seed: 10
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 50
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0002	1	6.0458
11.3955	0.0082	50	4.9855
10.3443	0.0164	100	5.0261
9.9611	0.0246	150	4.4048
9.9443	0.0328	200	4.2596
10.0276	0.0410	250	4.1043
10.0606	0.0492	300	4.1049
10.4285	0.0573	350	4.0138
9.4125	0.0655	400	3.9669
9.4526	0.0737	450	3.9476
9.6802	0.0819	500	3.9434

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1