Transaminitis_L3_100steps_1e6rate_05beta_CSFTDPO – AI Model by tsavage68 | AlphaNeural AI

You can deploy this model and start earning money today!

tsavage68/
Transaminitis_L3_100steps_1e6rate_05beta_CSFTDPO

like

0

text-generation

generated_from_trainer

tsavage68/Transaminitis_L3_1000rate_1e7_SFT

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

Transaminitis_L3_100steps_1e6rate_05beta_CSFTDPO

This model is a fine-tuned version of tsavage68/Transaminitis_L3_1000rate_1e7_SFT on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.0005
Rewards/chosen: 4.5705
Rewards/rejected: -5.2643
Rewards/accuracies: 1.0
Rewards/margins: 9.8348
Logps/rejected: -29.0833
Logps/chosen: -9.3932
Logits/rejected: -0.9625
Logits/chosen: -0.9598

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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.7055	0.2	25	0.6877	-0.5880	-0.6760	0.5400	0.0880	-19.9066	-19.7102	-1.0695	-1.0683
0.687	0.4	50	0.5510	3.8005	3.0720	0.6000	0.7284	-12.4107	-10.9333	-1.0238	-1.0226
0.4531	0.6	75	0.1502	4.9713	-4.7070	0.9800	9.6782	-27.9686	-8.5917	-1.0329	-1.0318
0.55	0.8	100	0.0005	4.5705	-5.2643	1.0	9.8348	-29.0833	-9.3932	-0.9625	-0.9598

Framework versions

Transformers 4.40.2
Pytorch 2.0.0+cu117
Datasets 2.19.1
Tokenizers 0.19.1