llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-q_t-0.5-s_star-0.6/README.md

library_name, base_model, tags, datasets, model-index

library_name	base_model	tags	datasets	model-index
transformers	W-61/llama-3-8b-base-sft-hh-harmless-4xh200	alignment-handbook new-dpo generated_from_trainer	Anthropic/hh-rlhf	name results llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-q_t-0.5-s_star-0.6

llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-q_t-0.5-s_star-0.6

This model is a fine-tuned version of W-61/llama-3-8b-base-sft-hh-harmless-4xh200 on the Anthropic/hh-rlhf dataset. It achieves the following results on the evaluation set:

• Loss: 0.5318
• Fcm Dpo/beta: 0.0183
• Margin Dpo/margin Mean: 34.3262
• Margin Dpo/margin Std: 53.3636
• Logps/chosen: -139.1072
• Logps/rejected: -178.1229
• Logps/ref Chosen: -74.8595
• Logps/ref Rejected: -79.5490
• Logits/chosen: 0.6984
• Logits/rejected: 0.6510

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: 5e-07
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• total_eval_batch_size: 32
• 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
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Fcm Dpo/beta	Margin Dpo/margin Mean	Margin Dpo/margin Std	Logps/chosen	Logps/rejected	Logps/ref Chosen	Logps/ref Rejected	Logits/chosen	Logits/rejected
0.9546	0.3023	200	0.5605	0.3456	1.6518	2.9692	-78.8826	-85.2239	-74.8595	-79.5490	0.2179	0.1797
1.1411	0.6047	400	0.5378	0.0260	22.3130	35.3835	-110.4847	-137.4872	-74.8595	-79.5490	0.6215	0.5730
1.1307	0.9070	600	0.5318	0.0183	34.3262	53.3636	-139.1072	-178.1229	-74.8595	-79.5490	0.6984	0.6510

Framework versions

• Transformers 4.51.0
• Pytorch 2.3.1+cu121
• Datasets 2.21.0
• Tokenizers 0.21.4