english-telugu-colloquial-translator

This model is a fine-tuned version of unsloth/tinyllama-chat-bnb-4bit on the None dataset. It achieves the following results on the evaluation set:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 0.0003
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 8
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
lr_scheduler_warmup_steps: 2
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
14.6072	0.2	2	9.2992
14.7062	0.4	4	9.2992
14.7707	0.6	6	9.2992
14.7094	0.8	8	9.2992
14.722	1.0	10	9.2992
14.6891	1.2	12	9.2992
14.7003	1.4	14	9.2992
14.7556	1.6	16	9.2994
10.6826	1.8	18	9.2998
9.8697	2.0	20	9.3005
11.1377	2.2	22	9.3005
7.7884	2.4	24	9.3002
6.0632	2.6	26	9.2997
5.2584	2.8	28	9.2984
4.8989	3.0	30	9.2961
4.6178	3.2	32	9.2924
4.4081	3.4	34	9.2877
4.1896	3.6	36	9.2824
4.1095	3.8	38	9.2774
3.998	4.0	40	9.2726
3.9535	4.2	42	9.2671
3.9242	4.4	44	9.2624
3.9082	4.6	46	9.2579
3.8371	4.8	48	9.2548
3.8282	5.0	50	9.2529
3.7675	5.2	52	9.2508
3.8366	5.4	54	9.2504
3.8349	5.6	56	9.2502
3.76	5.8	58	9.2492
3.7513	6.0	60	9.2491
3.7074	6.2	62	9.2493
3.6912	6.4	64	9.2487
3.7123	6.6	66	9.2474
3.6827	6.8	68	9.2481
3.68	7.0	70	9.2487
3.664	7.2	72	9.2483
3.7076	7.4	74	9.2482
3.6248	7.6	76	9.2486
3.6861	7.8	78	9.2492
3.6525	8.0	80	9.2495
3.6564	8.2	82	9.2499
3.6094	8.4	84	9.2502
3.5723	8.6	86	9.2505
3.5466	8.8	88	9.2506
3.5512	9.0	90	9.2508
3.5659	9.2	92	9.2508
3.6072	9.4	94	9.2510
3.5422	9.6	96	9.2511
3.5725	9.8	98	9.2514
3.5284	10.0	100	9.2516