This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct on the simonycl/llama3.1-ultrafeedback-annotate-armorm dataset. It achieves the following results on the evaluation set:

Loss: 0.3837
Rewards/chosen: -3.2511
Rewards/rejected: -5.1202
Rewards/accuracies: 0.8644
Rewards/margins: 1.8691
Logps/rejected: -797.6878
Logps/chosen: -602.0981
Logits/rejected: -1.3603
Logits/chosen: -1.3921

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: 1
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 32
total_train_batch_size: 128
total_eval_batch_size: 4
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
0.4269	0.8444	400	0.3837	-3.2511	-5.1202	0.8644	1.8691	-797.6878	-602.0981	-1.3603	-1.3921

Framework versions

Transformers 4.44.2
Pytorch 2.4.0+cu121
Datasets 2.21.0
Tokenizers 0.19.1

simonycl/llama-3.1-8b-instruct-ultrafeedback-armorm

llama-3.1-8b-instruct-armorm

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

simonycl/
llama-3.1-8b-instruct-ultrafeedback-armorm