This model is a fine-tuned version of YYYYYYibo/nash_rank4_all_iter_2 on the updated and the original datasets. It achieves the following results on the evaluation set:

Loss: 0.5483
Rewards/chosen: -0.5406
Rewards/rejected: -1.1226
Rewards/accuracies: 0.6960
Rewards/margins: 0.5820
Logps/rejected: -399.9344
Logps/chosen: -360.6637
Logits/rejected: 0.5949
Logits/chosen: -0.0563

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: 2
eval_batch_size: 2
seed: 42
distributed_type: multi-GPU
num_devices: 4
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
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
0.5359	0.49	100	0.5606	-0.4171	-0.9562	0.7000	0.5391	-383.2901	-348.3135	0.5653	-0.0477
0.5121	0.98	200	0.5483	-0.5406	-1.1226	0.6960	0.5820	-399.9344	-360.6637	0.5949	-0.0563

Framework versions

PEFT 0.7.1
Transformers 4.36.2
Pytorch 2.1.2+cu121
Datasets 2.14.6
Tokenizers 0.15.2

YYYYYYibo/nash_rank4_all_iter_3

nash_rank4_all_iter_3

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

YYYYYYibo/
nash_rank4_all_iter_3