learning_rate: 5e-05
train_batch_size: 4
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Chrf++	Gen Len
11.1882	1.0	2	1.2910	22.1085	106.875
11.1882	2.0	4	1.2869	22.1864	106.875
11.1882	3.0	6	1.2815	22.1797	106.5
11.1882	4.0	8	1.2740	23.4305	104.125
11.1882	5.0	10	1.2650	23.2799	104.75
11.1882	6.0	12	1.2534	23.6366	102.1875
11.1882	7.0	14	1.2406	23.6252	101.125
11.1882	8.0	16	1.2260	24.526	99.5
11.1882	9.0	18	1.2110	21.5763	115.375
11.1882	10.0	20	1.1946	22.3542	104.0
11.1882	11.0	22	1.1779	22.5292	102.375
11.1882	12.0	24	1.1595	23.5998	98.25
11.1882	13.0	26	1.1388	22.2512	121.0
11.1882	14.0	28	1.1174	24.4653	111.4375
11.1882	15.0	30	1.0980	23.908	125.9375
11.1882	16.0	32	1.0815	28.3564	87.6875
11.1882	17.0	34	1.0591	27.7864	89.9375
11.1882	18.0	36	1.0378	26.9864	83.25
11.1882	19.0	38	1.0164	27.4691	89.8125
11.1882	20.0	40	0.9979	27.7496	82.0625
11.1882	21.0	42	0.9780	24.8068	118.9375
11.1882	22.0	44	0.9572	27.4912	81.6875
11.1882	23.0	46	0.9378	26.9216	86.875
11.1882	24.0	48	0.9188	27.5656	84.6875
11.1882	25.0	50	0.9021	28.4866	75.0
11.1882	26.0	52	0.8852	29.8513	72.625
11.1882	27.0	54	0.8693	30.6218	71.1875
11.1882	28.0	56	0.8543	29.7969	69.9375
11.1882	29.0	58	0.8391	31.506	66.25
11.1882	30.0	60	0.8277	32.0042	62.3125
11.1882	31.0	62	0.8157	31.2593	64.0625
11.1882	32.0	64	0.8049	31.531	63.6875
11.1882	33.0	66	0.7947	32.4129	60.375
11.1882	34.0	68	0.7856	32.687	60.625
11.1882	35.0	70	0.7772	32.6299	60.5625
11.1882	36.0	72	0.7681	33.8175	60.875
11.1882	37.0	74	0.7610	34.0595	60.4375
11.1882	38.0	76	0.7556	35.2929	60.3125
11.1882	39.0	78	0.7496	37.2235	60.3125
11.1882	40.0	80	0.7446	37.6686	61.5
11.1882	41.0	82	0.7393	37.7114	61.3125
11.1882	42.0	84	0.7334	37.0134	61.625
11.1882	43.0	86	0.7296	37.7652	61.5
11.1882	44.0	88	0.7255	38.3566	62.4375
11.1882	45.0	90	0.7203	38.5978	60.5
11.1882	46.0	92	0.7164	39.6195	63.4375
11.1882	47.0	94	0.7114	39.9612	62.875
11.1882	48.0	96	0.7098	38.0797	63.125
11.1882	49.0	98	0.7070	38.6895	63.0
7.9761	50.0	100	0.7066	38.5245	62.6875
7.9761	51.0	102	0.7053	38.8923	63.125
7.9761	52.0	104	0.7031	38.2762	62.25
7.9761	53.0	106	0.7038	39.6654	63.1875
7.9761	54.0	108	0.7011	40.1876	62.9375
7.9761	55.0	110	0.6991	40.4171	65.6875
7.9761	56.0	112	0.6994	40.6954	64.875
7.9761	57.0	114	0.6971	38.8604	64.375
7.9761	58.0	116	0.6991	40.4291	64.625
7.9761	59.0	118	0.6975	41.3969	63.5625
7.9761	60.0	120	0.6990	42.015	64.125
7.9761	61.0	122	0.6999	41.1492	64.875
7.9761	62.0	124	0.6989	42.202	63.25
7.9761	63.0	126	0.7039	43.3173	65.125
7.9761	64.0	128	0.7056	43.8743	65.375
7.9761	65.0	130	0.7050	43.5723	64.75
7.9761	66.0	132	0.7097	43.9649	66.0625
7.9761	67.0	134	0.7118	43.4948	66.125
7.9761	68.0	136	0.7132	42.7592	65.125
7.9761	69.0	138	0.7185	42.6435	65.3125
7.9761	70.0	140	0.7216	42.7388	65.8125
7.9761	71.0	142	0.7215	43.5189	66.9375
7.9761	72.0	144	0.7299	44.6636	66.25
7.9761	73.0	146	0.7325	45.32	67.1875
7.9761	74.0	148	0.7386	42.9918	65.4375
7.9761	75.0	150	0.7494	44.7388	65.875
7.9761	76.0	152	0.7516	43.8916	65.5625
7.9761	77.0	154	0.7618	42.5172	65.875
7.9761	78.0	156	0.7658	43.297	65.875
7.9761	79.0	158	0.7704	44.4019	64.75
7.9761	80.0	160	0.7764	43.5389	64.75
7.9761	81.0	162	0.7847	43.4305	65.125
7.9761	82.0	164	0.7897	43.1141	65.75
7.9761	83.0	166	0.7940	43.2121	63.9375
7.9761	84.0	168	0.7995	45.061	65.25
7.9761	85.0	170	0.8153	42.7463	65.3125
7.9761	86.0	172	0.8239	44.1277	64.5
7.9761	87.0	174	0.8301	42.9544	64.875
7.9761	88.0	176	0.8376	41.7299	64.25

Framework versions

Transformers 5.12.1
Pytorch 2.6.0+cu124
Datasets 5.0.0
Tokenizers 0.22.2

Tsedeniya/finetuned_glosslm_pan_joint

finetuned_glosslm_pan_joint

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Tsedeniya/
finetuned_glosslm_pan_joint