O0506TEST

This model is a fine-tuned version of allenai/OLMo-1B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: -47.2856

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_steps: 100
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
-11.8749	0.09	10	-14.0218
-14.1305	0.18	20	-14.2468
-14.3704	0.27	30	-14.5481
-14.7299	0.36	40	-14.9938
-15.2303	0.45	50	-15.5855
-15.8744	0.54	60	-16.3445
-16.7251	0.63	70	-17.2990
-17.7631	0.73	80	-18.4397
-19.0466	0.82	90	-19.9335
-20.6521	0.91	100	-21.7543
-22.5524	1.0	110	-23.8121
-24.7052	1.09	120	-26.0076
-26.5875	1.18	130	-28.2672
-28.9769	1.27	140	-30.5118
-31.3719	1.36	150	-32.7197
-33.203	1.45	160	-34.8411
-35.4209	1.54	170	-36.8347
-37.2307	1.63	180	-38.6589
-38.8488	1.72	190	-37.7034
-40.0554	1.81	200	-41.4002
-41.1722	1.9	210	-42.2517
-42.2511	1.99	220	-43.2108
-43.6435	2.08	230	-44.1210
-44.0969	2.18	240	-44.9382
-44.1582	2.27	250	-45.6027
-45.5951	2.36	260	-46.1312
-46.2215	2.45	270	-46.5436
-45.4689	2.54	280	-46.8401
-46.9295	2.63	290	-47.0472
-46.8051	2.72	300	-47.1790
-46.5686	2.81	310	-47.2487
-46.5529	2.9	320	-47.2782
-47.2225	2.99	330	-47.2856

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.2+cu121
• Datasets 2.14.6
• Tokenizers 0.14.0