This model is a fine-tuned version of tsavage68/UTI_M2_1000steps_1e5rate_SFT on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.0695
Rewards/chosen: 0.6224
Rewards/rejected: -8.0773
Rewards/accuracies: 0.9000
Rewards/margins: 8.6997
Logps/rejected: -71.0904
Logps/chosen: -18.2199
Logits/rejected: -3.8351
Logits/chosen: -3.7548

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: 1e-06
train_batch_size: 2
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 4
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 100
training_steps: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.3884	0.3333	25	0.1489	0.7207	-2.4105	0.9000	3.1312	-52.2010	-17.8923	-3.8266	-3.7517
0.0182	0.6667	50	0.0695	0.6224	-8.0773	0.9000	8.6997	-71.0904	-18.2199	-3.8351	-3.7548

Framework versions

Transformers 4.41.2
Pytorch 2.0.0+cu117
Datasets 2.19.2
Tokenizers 0.19.1

tsavage68/UTI_M2_50steps_1e6rate_03beta_CSFTDPO

UTI_M2_50steps_1e6rate_03beta_CSFTDPO

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

tsavage68/
UTI_M2_50steps_1e6rate_03beta_CSFTDPO