learning_rate: 5e-06
train_batch_size: 32
eval_batch_size: 64
seed: 1024
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
5.2807	0.4213	1000	5.2465
4.7471	0.8425	2000	4.7511
4.2835	1.2637	3000	4.2718
3.9194	1.6850	4000	3.9048
3.6117	2.1062	5000	3.6040
3.4361	2.5274	6000	3.4308
3.3321	2.9487	7000	3.3178
3.2333	3.3699	8000	3.2541
3.1897	3.7911	9000	3.2000
3.1193	4.2123	10000	3.1608
3.0948	4.6336	11000	3.1265
3.0285	5.0548	12000	3.1035
3.0216	5.4760	13000	3.0836
3.0129	5.8973	14000	3.0615
2.9641	6.3185	15000	3.0523
2.964	6.7397	16000	3.0365
2.9113	7.1609	17000	3.0294
2.9193	7.5822	18000	3.0166
2.9193	8.0034	19000	3.0111
2.8916	8.4246	20000	3.0001
2.8872	8.8459	21000	2.9887
2.8482	9.2671	22000	2.9898
2.8533	9.6883	23000	2.9811
2.8097	10.1095	24000	2.9783
2.8264	10.5308	25000	2.9736
2.8303	10.9520	26000	2.9632
2.7991	11.3732	27000	2.9642
2.8044	11.7945	28000	2.9578
2.7656	12.2157	29000	2.9581
2.7798	12.6369	30000	2.9522
2.7379	13.0581	31000	2.9546
2.7511	13.4794	32000	2.9499
2.764	13.9006	33000	2.9440
2.7354	14.3218	34000	2.9468
2.7397	14.7431	35000	2.9433
2.7094	15.1643	36000	2.9434
2.7194	15.5855	37000	2.9387
2.7168	16.0067	38000	2.9427
2.7009	16.4280	39000	2.9405
2.7037	16.8492	40000	2.9355
2.6828	17.2704	41000	2.9381
2.6882	17.6917	42000	2.9349
2.6663	18.1129	43000	2.9375
2.6719	18.5341	44000	2.9352
2.6724	18.9554	45000	2.9338
2.6634	19.3766	46000	2.9349
2.6689	19.7979	47000	2.9343

kanishka/opt-babylm1_no_non_preemption_laugh-max_control_seed-1024_5e-6

opt-babylm1_no_non_preemption_laugh-max_control_seed-1024_5e-6