This model is a fine-tuned version of W-61/llama-3-8b-base-sft-ultrachat-8xh200 on the HuggingFaceH4/ultrafeedback_binarized dataset. It achieves the following results on the evaluation set:

Loss: 0.3504
Rewards/chosen: -0.2669
Rewards/rejected: -1.9046
Rewards/accuracies: 0.7460
Rewards/margins: 1.6376
Logps/rejected: -286.3951
Logps/chosen: -300.5437
Logits/rejected: -0.7660
Logits/chosen: -0.7842

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: 4
eval_batch_size: 4
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 4
total_train_batch_size: 128
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	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
1.5682	0.4188	200	0.4018	-0.2422	-1.5310	0.7540	1.2888	-282.6593	-300.2964	-0.7831	-0.8001
1.3853	0.8377	400	0.3504	-0.2669	-1.9046	0.7460	1.6376	-286.3951	-300.5437	-0.7660	-0.7842

Framework versions

Transformers 4.51.0
Pytorch 2.3.1+cu121
Datasets 2.21.0
Tokenizers 0.21.4

jackf857/llama-3-8b-base-robust-dpo-ultrafeedback-8xh200

llama-3-8b-base-robust-dpo-ultrafeedback-8xh200

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

jackf857/
llama-3-8b-base-robust-dpo-ultrafeedback-8xh200