exceptions_exp2_swap_take_to_drop_2128

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

• Loss: 3.5580
• Accuracy: 0.3695

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: 2128
• 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
4.829	0.2911	1000	0.2549	4.7516
4.3354	0.5822	2000	0.2993	4.2876
4.147	0.8733	3000	0.3157	4.0962
3.9906	1.1642	4000	0.3255	3.9941
3.9338	1.4553	5000	0.3319	3.9181
3.8809	1.7464	6000	0.3367	3.8595
3.7521	2.0373	7000	0.3414	3.8178
3.7646	2.3284	8000	0.3445	3.7855
3.7479	2.6195	9000	0.3472	3.7563
3.731	2.9106	10000	0.3495	3.7293
3.6314	3.2014	11000	0.3514	3.7177
3.6512	3.4925	12000	0.3530	3.6995
3.6358	3.7837	13000	0.3547	3.6827
3.5343	4.0745	14000	0.3563	3.6767
3.5705	4.3656	15000	0.3571	3.6637
3.5819	4.6567	16000	0.3585	3.6490
3.5651	4.9478	17000	0.3594	3.6356
3.504	5.2387	18000	0.3602	3.6370
3.5158	5.5298	19000	0.3612	3.6266
3.5296	5.8209	20000	0.3617	3.6152
3.4522	6.1121	21000	0.3620	3.6233
3.4725	6.4032	22000	0.3630	3.6138
3.4922	6.6943	23000	0.3637	3.6031
3.5007	6.9854	24000	0.3649	3.5906
3.4275	7.2763	25000	0.3646	3.6003
3.4465	7.5674	26000	0.3656	3.5898
3.4704	7.8585	27000	0.3659	3.5827
3.3769	8.1493	28000	0.3660	3.5926
3.4175	8.4404	29000	0.3665	3.5867
3.4279	8.7315	30000	0.3672	3.5748
3.3299	9.0224	31000	0.3671	3.5813
3.3795	9.3135	32000	0.3671	3.5809
3.3921	9.6046	33000	0.3674	3.5738
3.4213	9.8957	34000	0.3686	3.5627
3.3333	10.1866	35000	0.3683	3.5709
3.3679	10.4777	36000	0.3685	3.5674
3.3889	10.7688	37000	0.3694	3.5587
3.2762	11.0597	38000	0.3692	3.5703
3.3422	11.3508	39000	0.3692	3.5667
3.3645	11.6419	40000	0.3695	3.5580
3.3777	11.9330	41000	0.3704	3.5492
3.3127	12.2239	42000	0.3699	3.5646
3.3324	12.5150	43000	0.3703	3.5546
3.35	12.8061	44000	0.3708	3.5501
3.2664	13.0969	45000	0.3705	3.5623
3.3057	13.3880	46000	0.3705	3.5554
3.3299	13.6791	47000	0.3710	3.5501
3.3344	13.9702	48000	0.3715	3.5411
3.2717	14.2611	49000	0.3708	3.5613
3.3047	14.5522	50000	0.3711	3.5541
3.3198	14.8433	51000	0.3717	3.5440
3.2314	15.1342	52000	0.3710	3.5598
3.288	15.4253	53000	0.3713	3.5509
3.292	15.7164	54000	0.3720	3.5418
3.2576	16.0073	55000	0.3715	3.5504
3.2485	16.2984	56000	0.3719	3.5515
3.2742	16.5895	57000	0.3722	3.5468
3.3017	16.8806	58000	0.3726	3.5375
3.2229	17.1715	59000	0.3720	3.5524
3.2657	17.4626	60000	0.3724	3.5456
3.2926	17.7537	61000	0.3728	3.5415
3.194	18.0445	62000	0.3722	3.5531
3.2321	18.3356	63000	0.3725	3.5496
3.2589	18.6267	64000	0.3726	3.5415
3.2746	18.9179	65000	0.3733	3.5348
3.2104	19.2087	66000	0.3726	3.5512
3.2527	19.4998	67000	0.3729	3.5433
3.2465	19.7909	68000	0.3734	3.5340
3.1742	20.0818	69000	0.3728	3.5496
3.2113	20.3729	70000	0.3730	3.5481
3.2383	20.6640	71000	0.3736	3.5364
3.2497	20.9551	72000	0.3738	3.5342
3.1931	21.2460	73000	0.3730	3.5484
3.2141	21.5371	74000	0.3735	3.5425
3.2294	21.8282	75000	0.3740	3.5341
3.1546	22.1191	76000	0.3736	3.5517
3.1986	22.4102	77000	0.3735	3.5471
3.2226	22.7013	78000	0.3738	3.5435
3.2436	22.9924	79000	0.3746	3.5277
3.1827	23.2832	80000	0.3734	3.5515
3.1975	23.5743	81000	0.3739	3.5425
3.2275	23.8655	82000	0.3741	3.5339
3.1521	24.1563	83000	0.3736	3.5519
3.1829	24.4474	84000	0.3735	3.5469
3.2103	24.7385	85000	0.3743	3.5389
3.1168	25.0294	86000	0.3738	3.5512
3.1637	25.3205	87000	0.3736	3.5502
3.1768	25.6116	88000	0.3740	3.5454
3.2057	25.9027	89000	0.3748	3.5337
3.1333	26.1936	90000	0.3736	3.5514
3.1644	26.4844	91000	3.5537	0.3738
3.1785	26.7755	92000	3.5467	0.3742
3.115	27.0667	93000	3.5535	0.3737
3.1541	27.3578	94000	3.5506	0.3742
3.1689	27.6489	95000	3.5420	0.3745
3.1867	27.9400	96000	3.5365	0.3751
3.1251	28.2308	97000	3.5566	0.3739
3.1487	28.5219	98000	3.5452	0.3747
3.1661	28.8131	99000	3.5379	0.3750
3.1053	29.1039	100000	3.5556	0.3738
3.1238	29.3950	101000	3.5503	0.3745
3.1671	29.6861	102000	3.5434	0.3748
3.1642	29.9772	103000	3.5369	0.3751
3.1046	30.2681	104000	3.5547	0.3743
3.1332	30.5592	105000	3.5464	0.3748
3.1583	30.8503	106000	3.5396	0.3752
3.0952	31.1412	107000	3.5569	0.3741
3.1241	31.4323	108000	3.5512	0.3748
3.1344	31.7234	109000	3.5416	0.3752
3.0589	32.0143	110000	3.5506	0.3748

Framework versions

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