learning_rate: 3.6e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 100
num_epochs: 1
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
0.8495	0.0089	200	0.8262
0.8218	0.0178	400	0.8059
0.7938	0.0267	600	0.7957
0.7868	0.0356	800	0.7859
0.7820	0.0444	1000	0.7762
0.7834	0.0533	1200	0.7689
0.7560	0.0622	1400	0.7634
0.7560	0.0711	1600	0.7578
0.7204	0.08	1800	0.7520
0.6891	0.0889	2000	0.7507
0.7452	0.0978	2200	0.7544
0.8084	0.1067	2400	0.7617
0.8284	0.1156	2600	0.7859
0.9347	0.1244	2800	0.8551
1.5500	0.1333	3000	1.4804
1.6737	0.1422	3200	1.9633
2.8793	0.1511	3400	2.4767
4.3766	0.16	3600	4.4668
8.7282	0.1689	3800	8.8004
9.0311	0.1778	4000	8.9600
9.0768	0.1867	4200	8.9586
9.0761	0.1956	4400	8.9573
9.0046	0.2044	4600	8.9557
8.9676	0.2133	4800	8.9545
9.1755	0.2222	5000	8.9535
9.1615	0.2311	5200	8.9523
8.9258	0.24	5400	8.9518
9.1123	0.2489	5600	8.9500
8.9611	0.2578	5800	8.9490
8.9669	0.2667	6000	8.9479
9.0853	0.2756	6200	8.9467
8.9731	0.2844	6400	8.9454
9.0414	0.2933	6600	8.9442
9.0170	0.3022	6800	8.9436
9.0352	0.3111	7000	8.9423
9.0178	0.32	7200	8.9410
9.0483	0.3289	7400	8.9396
8.9148	0.3378	7600	8.9386
9.0972	0.3467	7800	8.9375
8.8510	0.3556	8000	8.9366
8.9616	0.3644	8200	8.9358
8.8572	0.3733	8400	8.9350
8.9840	0.3822	8600	8.9341
8.9124	0.3911	8800	8.9332
8.8301	0.4	9000	8.9309
9.0571	0.4089	9200	8.9314
9.0379	0.4178	9400	8.9302
9.0253	0.4267	9600	8.9282
9.0509	0.4356	9800	8.9264
9.0200	0.4444	10000	8.9251
8.9515	0.4533	10200	8.9243
8.9868	0.4622	10400	8.9260
8.8092	0.4711	10600	8.9227
8.9911	0.48	10800	8.9218
9.0521	0.4889	11000	8.9223
9.0731	0.4978	11200	8.9208
8.9302	0.5067	11400	8.9211
8.9997	0.5156	11600	8.9202
8.8510	0.5244	11800	8.9184
9.0046	0.5333	12000	8.9188

strongpear/Llama3.1-8B-Q_G_D1_D2_CoT_A-LAW-Instruct-r64

Llama3.1-8B-Q_G_D1_D2_CoT_A-LAW-Instruct-r64