zephyr-7b-uf-rlced-conifer-dpo-2e

This model is a fine-tuned version of alignment-handbook/zephyr-7b-sft-full on the data/uf_rlced_conifer dataset. It achieves the following results on the evaluation set:

• Loss: 0.2466
• Rewards/chosen: -4.8588
• Rewards/rejected: -13.2530
• Rewards/accuracies: 0.8966
• Rewards/margins: 8.3941
• Logps/rejected: -1735.4203
• Logps/chosen: -870.4680
• Logits/rejected: 4.3075
• Logits/chosen: 1.8380

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: 8
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• total_eval_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.1587	1.3879	1000	0.2471	-3.9472	-12.0338	0.8910	8.0866	-1613.5016	-779.3055	4.5606	2.4398

Framework versions

• Transformers 4.44.1
• Pytorch 2.1.2+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1