exceptions_exp2_swap_take_to_hit_5039

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

• Loss: 3.5570
• Accuracy: 0.3698

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: 0.0006
• train_batch_size: 16
• eval_batch_size: 16
• seed: 5039
• gradient_accumulation_steps: 5
• total_train_batch_size: 80
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 50.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
4.8267	0.2911	1000	0.2556	4.7471
4.335	0.5822	2000	0.2992	4.2812
4.1422	0.8733	3000	0.3159	4.0929
3.9853	1.1642	4000	0.3253	3.9873
3.9251	1.4553	5000	0.3322	3.9105
3.8682	1.7464	6000	0.3375	3.8539
3.7454	2.0373	7000	0.3416	3.8124
3.7501	2.3284	8000	0.3450	3.7811
3.7337	2.6195	9000	0.3473	3.7509
3.7241	2.9106	10000	0.3500	3.7274
3.6349	3.2014	11000	0.3520	3.7130
3.6473	3.4925	12000	0.3537	3.6923
3.641	3.7837	13000	0.3550	3.6779
3.5416	4.0745	14000	0.3564	3.6689
3.5641	4.3656	15000	0.3573	3.6569
3.5772	4.6567	16000	0.3588	3.6424
3.5842	4.9478	17000	0.3601	3.6284
3.5059	5.2387	18000	0.3607	3.6320
3.5227	5.5298	19000	0.3614	3.6231
3.5259	5.8209	20000	0.3625	3.6127
3.4557	6.1121	21000	0.3619	3.6238
3.4692	6.4032	22000	0.3632	3.6104
3.4815	6.6943	23000	0.3640	3.6007
3.4941	6.9854	24000	0.3648	3.5885
3.4138	7.2763	25000	0.3649	3.5998
3.4522	7.5674	26000	0.3654	3.5897
3.4632	7.8585	27000	0.3666	3.5792
3.3757	8.1493	28000	0.3663	3.5883
3.4123	8.4404	29000	0.3670	3.5789
3.4219	8.7315	30000	0.3674	3.5708
3.3216	9.0224	31000	0.3672	3.5786
3.3661	9.3135	32000	0.3678	3.5772
3.3982	9.6046	33000	0.3684	3.5668
3.4211	9.8957	34000	0.3687	3.5583
3.3213	10.1866	35000	0.3681	3.5751
3.3611	10.4777	36000	0.3690	3.5656
3.3795	10.7688	37000	0.3696	3.5571
3.2787	11.0597	38000	0.3695	3.5660
3.3384	11.3508	39000	0.3694	3.5648
3.358	11.6419	40000	0.3698	3.5570
3.3667	11.9330	41000	0.3708	3.5489
3.3017	12.2239	42000	0.3699	3.5628
3.3324	12.5150	43000	0.3708	3.5560
3.3493	12.8061	44000	0.3711	3.5466
3.2535	13.0969	45000	0.3708	3.5593
3.2969	13.3880	46000	0.3711	3.5569
3.3297	13.6791	47000	0.3715	3.5479
3.3497	13.9702	48000	0.3720	3.5400
3.2808	14.2611	49000	0.3713	3.5549
3.313	14.5522	50000	0.3719	3.5444
3.3258	14.8433	51000	0.3718	3.5429
3.2322	15.1342	52000	0.3715	3.5568
3.2845	15.4253	53000	0.3721	3.5448
3.2968	15.7164	54000	0.3725	3.5418
3.2486	16.0073	55000	0.3721	3.5474
3.2576	16.2984	56000	0.3724	3.5514
3.2762	16.5895	57000	0.3725	3.5426
3.2808	16.8806	58000	0.3732	3.5349
3.2286	17.1715	59000	0.3726	3.5512
3.249	17.4626	60000	0.3728	3.5443
3.2756	17.7537	61000	0.3734	3.5346
3.1878	18.0445	62000	0.3729	3.5485
3.2435	18.3356	63000	0.3728	3.5461
3.2444	18.6267	64000	0.3735	3.5416
3.2737	18.9179	65000	0.3739	3.5312
3.2172	19.2087	66000	0.3733	3.5472
3.2437	19.4998	67000	0.3736	3.5403
3.257	19.7909	68000	0.3741	3.5346
3.166	20.0818	69000	0.3734	3.5492
3.2217	20.3729	70000	0.3737	3.5441
3.2449	20.6640	71000	0.3738	3.5375
3.2546	20.9551	72000	0.3744	3.5299
3.1902	21.2460	73000	0.3735	3.5467
3.2182	21.5371	74000	0.3742	3.5405
3.2294	21.8282	75000	0.3745	3.5300
3.1699	22.1191	76000	0.3738	3.5485
3.1891	22.4102	77000	0.3742	3.5421
3.2212	22.7013	78000	0.3742	3.5400
3.2507	22.9924	79000	0.3750	3.5283
3.1879	23.2832	80000	0.3740	3.5454
3.2108	23.5743	81000	0.3744	3.5346
3.2218	23.8655	82000	0.3748	3.5333
3.1481	24.1563	83000	0.3741	3.5494
3.1827	24.4474	84000	0.3746	3.5413
3.2085	24.7385	85000	0.3751	3.5319
3.1125	25.0294	86000	0.3744	3.5471
3.1476	25.3205	87000	0.3741	3.5468
3.1828	25.6116	88000	0.3748	3.5356
3.1948	25.9027	89000	0.3752	3.5324
3.1446	26.1936	90000	0.3743	3.5476
3.1514	26.4844	91000	3.5471	0.3744
3.1652	26.7755	92000	3.5409	0.3748
3.0971	27.0667	93000	3.5543	0.3742
3.1488	27.3578	94000	3.5456	0.3746
3.1608	27.6489	95000	3.5400	0.3749
3.1884	27.9400	96000	3.5322	0.3755
3.1219	28.2308	97000	3.5495	0.3748
3.139	28.5219	98000	3.5403	0.3752
3.1504	28.8131	99000	3.5349	0.3753
3.0924	29.1039	100000	3.5472	0.3747
3.1364	29.3950	101000	3.5444	0.3748
3.1578	29.6861	102000	3.5361	0.3755
3.1669	29.9772	103000	3.5269	0.3759
3.1157	30.2681	104000	3.5489	0.3747
3.1444	30.5592	105000	3.5353	0.3758
3.1564	30.8503	106000	3.5342	0.3757
3.0836	31.1412	107000	3.5490	0.3752
3.1177	31.4323	108000	3.5441	0.3751
3.13	31.7234	109000	3.5346	0.3758
3.0624	32.0143	110000	3.5475	0.3754
3.0914	32.3054	111000	3.5468	0.3752
3.1166	32.5965	112000	3.5399	0.3756
3.1436	32.8876	113000	3.5346	0.3763
3.0634	33.1784	114000	3.5488	0.3756
3.0987	33.4696	115000	3.5428	0.3757
3.1201	33.7607	116000	3.5402	0.3758
3.0471	34.0515	117000	3.5495	0.3756
3.082	34.3426	118000	3.5463	0.3756
3.0968	34.6337	119000	3.5403	0.3760
3.1116	34.9248	120000	3.5352	0.3764
3.0695	35.2157	121000	3.5509	0.3756
3.0986	35.5068	122000	3.5445	0.3761
3.121	35.7979	123000	3.5374	0.3761

Framework versions

• Transformers 4.55.2
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4