learning_rate: 0.0003
train_batch_size: 256
eval_batch_size: 16
seed: 42
distributed_type: multi-GPU
gradient_accumulation_steps: 2
total_train_batch_size: 512
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-07 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 1

Training Loss	Epoch	Step	Validation Loss
2.191	0.0156	1000	0.3491
2.1395	0.0312	2000	0.3458
2.0383	0.0468	3000	0.3613
1.8148	0.0624	4000	0.3705
1.8758	0.0779	5000	0.3902
2.1148	0.0935	6000	0.4127
2.0859	0.1091	7000	0.4094
2.1773	0.1247	8000	0.3933
2.4336	0.1403	9000	0.3797
1.9531	0.1559	10000	0.3690
2.0906	0.1715	11000	0.3623
2.1289	0.1871	12000	0.3559
1.9367	0.2026	13000	0.3507
2.182	0.2182	14000	0.3500
2.0492	0.2338	15000	0.3444
2.0281	0.2494	16000	0.3373
2.0477	0.2650	17000	0.3357
1.7992	0.2806	18000	0.3338
1.9984	0.2962	19000	0.3273
1.7523	0.3118	20000	0.3214
1.7953	0.3273	21000	0.3205
1.507	0.3429	22000	0.3134
1.8336	0.3585	23000	0.3155
1.5641	0.3741	24000	0.3127
1.6559	0.3897	25000	0.3091
1.7469	0.4053	26000	0.3036
1.7961	0.4209	27000	0.3042
1.7545	0.4365	28000	0.3023
1.793	0.4520	29000	0.2982
1.6129	0.4676	30000	0.2948
1.4312	0.4832	31000	0.2962
1.6621	0.4988	32000	0.2943
1.6121	0.5144	33000	0.2869
1.4781	0.5300	34000	0.2868
1.4285	0.5456	35000	0.2857
1.2082	0.5612	36000	0.2855
1.427	0.5767	37000	0.2845
1.359	0.5923	38000	0.2831
1.3754	0.6079	39000	0.2798
1.5141	0.6235	40000	0.2773
1.2621	0.6391	41000	0.2763
1.3496	0.6547	42000	0.2793
1.1902	0.6703	43000	0.2773
1.1859	0.6859	44000	0.2747
1.3562	0.7014	45000	0.2769
1.2703	0.7170	46000	0.2739
1.323	0.7326	47000	0.2754
1.1734	0.7482	48000	0.2756
1.1773	0.7638	49000	0.2747
1.0867	0.7794	50000	0.2771
1.0625	0.7950	51000	0.2739
1.1129	0.8106	52000	0.2756
1.1064	0.8261	53000	0.2742
1.1336	0.8417	54000	0.2773
1.2156	0.8573	55000	0.2747
0.948	0.8729	56000	0.2761
1.0961	0.8885	57000	0.2765
0.9602	0.9041	58000	0.2754
0.8793	0.9197	59000	0.2751
1.3434	0.9353	60000	0.2745
1.0738	0.9509	61000	0.2752
1.1676	0.9664	62000	0.2752
0.8816	0.9820	63000	0.2753
1.1309	0.9976	64000	0.2751

ferrazzipietro/unsup-gemma-3-1b-it-datav3

unsup-gemma-3-1b-it-datav3