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: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.1809	0.2667	50	0.0764	0.2775	-2.3601	1.0	2.6376	-94.4829	-33.9523	-2.2243	-2.2318
0.0001	0.5333	100	0.0001	0.5624	-9.1803	1.0	9.7427	-162.6846	-31.1030	-2.0967	-2.1493
0.0	0.8	150	0.0000	0.6644	-9.8478	1.0	10.5122	-169.3592	-30.0830	-2.0886	-2.1437
0.0	1.0667	200	0.0000	0.7073	-10.2460	1.0	10.9533	-173.3410	-29.6539	-2.0839	-2.1405
0.0	1.3333	250	0.0000	0.7518	-10.5229	1.0	11.2746	-176.1101	-29.2095	-2.0800	-2.1377
0.0	1.6	300	0.0000	0.7875	-10.7312	1.0	11.5187	-178.1934	-28.8516	-2.0768	-2.1352
0.0	1.8667	350	0.0000	0.8183	-10.9114	1.0	11.7297	-179.9959	-28.5445	-2.0748	-2.1340
0.0	2.1333	400	0.0000	0.8479	-11.0550	1.0	11.9029	-181.4319	-28.2485	-2.0722	-2.1321
0.0	2.4	450	0.0000	0.8755	-11.1725	1.0	12.0480	-182.6062	-27.9716	-2.0707	-2.1309
0.0	2.6667	500	0.0000	0.8956	-11.2535	1.0	12.1491	-183.4164	-27.7715	-2.0697	-2.1303
0.0	2.9333	550	0.0000	0.9100	-11.3169	1.0	12.2269	-184.0507	-27.6270	-2.0687	-2.1297
0.0	3.2	600	0.0000	0.9219	-11.3489	1.0	12.2708	-184.3704	-27.5080	-2.0686	-2.1296
0.0	3.4667	650	0.0000	0.9289	-11.3756	1.0	12.3045	-184.6379	-27.4380	-2.0678	-2.1289
0.0	3.7333	700	0.0000	0.9343	-11.3811	1.0	12.3154	-184.6925	-27.3842	-2.0675	-2.1286
0.0	4.0	750	0.0000	0.9296	-11.3865	1.0	12.3160	-184.7462	-27.4315	-2.0664	-2.1276
0.0	4.2667	800	0.0000	0.9324	-11.3857	1.0	12.3182	-184.7389	-27.4029	-2.0672	-2.1284
0.0	4.5333	850	0.0000	0.9292	-11.3886	1.0	12.3178	-184.7675	-27.4349	-2.0668	-2.1279
0.0	4.8	900	0.0000	0.9312	-11.3911	1.0	12.3223	-184.7923	-27.4146	-2.0667	-2.1278
0.0	5.0667	950	0.0000	0.9318	-11.3910	1.0	12.3228	-184.7916	-27.4095	-2.0667	-2.1278
0.0	5.3333	1000	0.0000	0.9318	-11.3910	1.0	12.3228	-184.7916	-27.4095	-2.0667	-2.1278

Framework versions

Transformers 4.42.4
Pytorch 2.0.0+cu117
Datasets 2.20.0
Tokenizers 0.19.1

tsavage68/Hyponatremia_M2_1000steps_1e7rate_01beta_DPO

Hyponatremia_M2_1000steps_1e7rate_01beta_DPO

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

tsavage68/
Hyponatremia_M2_1000steps_1e7rate_01beta_DPO