This model is a fine-tuned version of tsavage68/UTI_M2_1000steps_1e7rate_SFT on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.6931
Rewards/chosen: 0.0
Rewards/rejected: 0.0
Rewards/accuracies: 0.0
Rewards/margins: 0.0
Logps/rejected: 0.0
Logps/chosen: 0.0
Logits/rejected: -2.7147
Logits/chosen: -2.7147

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-08
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: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Logits/rejected	Logits/chosen
0.6931	0.3333	25	0.6931	-2.7147	-2.7147
0.6931	0.6667	50	0.6931	-2.7147	-2.7147
0.6931	1.0	75	0.6931	-2.7147	-2.7147
0.6931	1.3333	100	0.6931	-2.7147	-2.7147
0.6931	1.6667	125	0.6931	-2.7147	-2.7147
0.6931	2.0	150	0.6931	-2.7147	-2.7147
0.6931	2.3333	175	0.6931	-2.7147	-2.7147
0.6931	2.6667	200	0.6931	-2.7147	-2.7147
0.6931	3.0	225	0.6931	-2.7147	-2.7147
0.6931	3.3333	250	0.6931	-2.7147	-2.7147
0.6931	3.6667	275	0.6931	-2.7147	-2.7147
0.6931	4.0	300	0.6931	-2.7147	-2.7147
0.6931	4.3333	325	0.6931	-2.7147	-2.7147
0.6931	4.6667	350	0.6931	-2.7147	-2.7147
0.6931	5.0	375	0.6931	-2.7147	-2.7147
0.6931	5.3333	400	0.6931	-2.7147	-2.7147
0.6931	5.6667	425	0.6931	-2.7147	-2.7147
0.6931	6.0	450	0.6931	-2.7147	-2.7147
0.6931	6.3333	475	0.6931	-2.7147	-2.7147
0.6931	6.6667	500	0.6931	-2.7147	-2.7147
0.6931	7.0	525	0.6931	-2.7147	-2.7147
0.6931	7.3333	550	0.6931	-2.7147	-2.7147
0.6931	7.6667	575	0.6931	-2.7147	-2.7147
0.6931	8.0	600	0.6931	-2.7147	-2.7147
0.6931	8.3333	625	0.6931	-2.7147	-2.7147
0.6931	8.6667	650	0.6931	-2.7147	-2.7147
0.6931	9.0	675	0.6931	-2.7147	-2.7147
0.6931	9.3333	700	0.6931	-2.7147	-2.7147
0.6931	9.6667	725	0.6931	-2.7147	-2.7147
0.6931	10.0	750	0.6931	-2.7147	-2.7147
0.6931	10.3333	775	0.6931	-2.7147	-2.7147
0.6931	10.6667	800	0.6931	-2.7147	-2.7147
0.6931	11.0	825	0.6931	-2.7147	-2.7147
0.6931	11.3333	850	0.6931	-2.7147	-2.7147
0.6931	11.6667	875	0.6931	-2.7147	-2.7147
0.6931	12.0	900	0.6931	-2.7147	-2.7147
0.6931	12.3333	925	0.6931	-2.7147	-2.7147
0.6931	12.6667	950	0.6931	-2.7147	-2.7147
0.6931	13.0	975	0.6931	-2.7147	-2.7147
0.6931	13.3333	1000	0.6931	-2.7147	-2.7147

Framework versions

Transformers 4.41.2
Pytorch 2.0.0+cu117
Datasets 2.19.2
Tokenizers 0.19.1

tsavage68/UTI2_M2_1000steps_1e8rate_01beat_CSFTDPO

UTI2_M2_1000steps_1e8rate_01beat_CSFTDPO

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

tsavage68/
UTI2_M2_1000steps_1e8rate_01beat_CSFTDPO