learning_rate: 5e-06
train_batch_size: 32
eval_batch_size: 64
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 256
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.05
num_epochs: 20.0

Training Loss	Epoch	Step	Validation Loss
5.2723	0.4205	1000	5.2218
4.6886	0.8410	2000	4.6682
4.2073	1.2611	3000	4.2120
3.8487	1.6816	4000	3.8280
3.5449	2.1018	5000	3.5509
3.3994	2.5223	6000	3.3959
3.2966	2.9428	7000	3.2974
3.2142	3.3629	8000	3.2404
3.1702	3.7834	9000	3.1843
3.0998	4.2035	10000	3.1505
3.0899	4.6240	11000	3.1190
3.0183	5.0442	12000	3.0975
3.0174	5.4647	13000	3.0778
3.0087	5.8852	14000	3.0575
2.9599	6.3053	15000	3.0456
2.9565	6.7258	16000	3.0320
2.9094	7.1459	17000	3.0233
2.9145	7.5664	18000	3.0145
2.9163	7.9869	19000	2.9998
2.8827	8.4070	20000	2.9996
2.8816	8.8275	21000	2.9871
2.8464	9.2477	22000	2.9863
2.8514	9.6682	23000	2.9776
2.8015	10.0883	24000	2.9784
2.819	10.5088	25000	2.9699
2.8284	10.9293	26000	2.9610
2.7917	11.3494	27000	2.9677
2.8078	11.7699	28000	2.9591
2.7691	12.1901	29000	2.9596
2.7745	12.6106	30000	2.9549
2.74	13.0307	31000	2.9525
2.751	13.4512	32000	2.9514
2.7605	13.8717	33000	2.9454
2.7322	14.2918	34000	2.9468
2.7386	14.7123	35000	2.9430
2.7066	15.1325	36000	2.9445
2.7176	15.5530	37000	2.9416
2.7179	15.9735	38000	2.9370
2.6963	16.3936	39000	2.9421
2.7074	16.8141	40000	2.9373
2.6814	17.2342	41000	2.9396
2.683	17.6547	42000	2.9367
2.6688	18.0748	43000	2.9373
2.6683	18.4953	44000	2.9366
2.6711	18.9158	45000	2.9351
2.6618	19.3360	46000	2.9363
2.6572	19.7565	47000	2.9356

kanishka/opt-babylm1_more_preemption_laugh_seed-42_5e-6

opt-babylm1_more_preemption_laugh_seed-42_5e-6