Hyponatremia_L3_250steps_1e7rate_03beta_DPO – AI Model by tsavage68 | AlphaNeural AI

You can deploy this model and start earning money today!

tsavage68/
Hyponatremia_L3_250steps_1e7rate_03beta_DPO

like

0

text-generation

generated_from_trainer

tsavage68/Hyponatremia_L3_450steps_1e7rate_SFT

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

Hyponatremia_L3_250steps_1e7rate_03beta_DPO

This model is a fine-tuned version of tsavage68/Hyponatremia_L3_450steps_1e7rate_SFT on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.0003
Rewards/chosen: 1.6038
Rewards/rejected: -6.7115
Rewards/accuracies: 1.0
Rewards/margins: 8.3153
Logps/rejected: -61.7984
Logps/chosen: -17.3708
Logits/rejected: -1.0140
Logits/chosen: -0.9245

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-07
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: 250

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.5498	0.2667	50	0.4912	0.1121	-0.3443	1.0	0.4564	-40.5745	-22.3430	-1.0192	-0.9446
0.0327	0.5333	100	0.0171	1.0812	-3.0693	1.0	4.1505	-49.6580	-19.1129	-1.0134	-0.9308
0.0004	0.8	150	0.0004	1.5615	-6.4242	1.0	7.9857	-60.8409	-17.5118	-1.0138	-0.9245
0.0003	1.0667	200	0.0003	1.6017	-6.7112	1.0	8.3129	-61.7974	-17.3778	-1.0142	-0.9247
0.0003	1.3333	250	0.0003	1.6038	-6.7115	1.0	8.3153	-61.7984	-17.3708	-1.0140	-0.9245

Framework versions

Transformers 4.42.3
Pytorch 2.0.0+cu117
Datasets 2.20.0
Tokenizers 0.19.1