Hyponatremia_M2_300steps_1e8rate_05beta_DPO – AI Model by tsavage68 | AlphaNeural AI

You can deploy this model and start earning money today!

tsavage68/
Hyponatremia_M2_300steps_1e8rate_05beta_DPO

like

0

text-generation

generated_from_trainer

tsavage68/Hyponatremia_M2_150steps_1e6rate_SFT

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

Hyponatremia_M2_300steps_1e8rate_05beta_DPO

This model is a fine-tuned version of tsavage68/Hyponatremia_M2_150steps_1e6rate_SFT on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.6226
Rewards/chosen: 0.0235
Rewards/rejected: -0.1272
Rewards/accuracies: 0.9000
Rewards/margins: 0.1507
Logps/rejected: -71.1359
Logps/chosen: -36.6801
Logits/rejected: -2.2466
Logits/chosen: -2.2415

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: 300

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.6917	0.2667	50	0.7006	0.0051	0.0168	0.4500	-0.0118	-70.8478	-36.7169	-2.2476	-2.2422
0.6801	0.5333	100	0.6899	0.0048	-0.0060	0.5900	0.0108	-70.8935	-36.7174	-2.2467	-2.2416
0.6084	0.8	150	0.6309	0.0226	-0.1105	0.8300	0.1331	-71.1025	-36.6818	-2.2465	-2.2414
0.6008	1.0667	200	0.6238	0.0352	-0.1130	0.9000	0.1483	-71.1075	-36.6566	-2.2466	-2.2416
0.614	1.3333	250	0.6226	0.0235	-0.1272	0.9000	0.1507	-71.1359	-36.6801	-2.2466	-2.2415
0.617	1.6	300	0.6226	0.0235	-0.1272	0.9000	0.1507	-71.1359	-36.6801	-2.2466	-2.2415

Framework versions

Transformers 4.42.4
Pytorch 2.0.0+cu117
Datasets 2.20.0
Tokenizers 0.19.1