This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the pref_data dataset.
It achieves the following results on the evaluation set:
Loss: 0.8061
Rewards/chosen: -0.1849
Rewards/rejected: -0.4492
Rewards/accuracies: 0.9873
Rewards/margins: 0.2643
Logps/rejected: -4.4920
Logps/chosen: -1.8486
Logits/rejected: 1.5509
Logits/chosen: 1.8515
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-06
train_batch_size: 1
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 16
total_train_batch_size: 128
total_eval_batch_size: 8
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
