lamso-en-nllb-custom-token

kimbolingo/lamso-en-nllb-custom-token

lamso-en-nllb-custom-token is a machine learning model from kimbolingo. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as cc-by-nc-4.0.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

4% of all-time downloads

All-time downloads

149

Public

Parameters

615M

5 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors2.5 GB · 98%

At a glance

Library: transformers
License: cc-by-nc-4.0
Model type: m2m_100
Access: Public
Created: Dec 3, 2025
Updated: Dec 19, 2025
SHA: 1626e697

Base models

facebook/nllb-200-distilled-600M

Model card

From the Hugging Face model README

lamso-en-nllb-custom-token

This model is a fine-tuned version of facebook/nllb-200-distilled-600M on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.8319
Bleu: 36.0841

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: 3e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 100
num_epochs: 50
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
3.2992	1.0	290	3.0374	1.7746
2.9452	2.0	580	2.6708	3.1774
2.6351	3.0	870	2.4456	4.7579
2.3747	4.0	1160	2.2995	6.1581
2.1906	5.0	1450	2.1960	7.9135
2.1335	6.0	1740	2.1053	10.0148
2.0035	7.0	2030	2.0402	11.7725
1.8662	8.0	2320	1.9873	13.8398
1.7376	9.0	2610	1.9428	14.4504
1.7713	10.0	2900	1.9049	15.9470
1.6119	11.0	3190	1.8758	16.7400
1.3993	12.0	3480	1.8552	18.5520
1.4369	13.0	3770	1.8239	19.1417
1.3949	14.0	4060	1.8221	20.5723
1.3175	15.0	4350	1.8020	22.0797
1.239	16.0	4640	1.7916	23.0882
1.185	17.0	4930	1.7860	24.0988
1.1963	18.0	5220	1.7832	24.7259
1.0723	19.0	5510	1.7828	25.4883
1.0228	20.0	5800	1.7802	26.4792
0.9696	21.0	6090	1.7724	27.2210
1.0731	22.0	6380	1.7642	27.3666
0.9944	23.0	6670	1.7857	27.9625
0.8621	24.0	6960	1.7797	28.9914
0.9625	25.0	7250	1.7784	29.4125
0.8799	26.0	7540	1.7822	30.0022
0.8268	27.0	7830	1.7763	30.4834
0.785	28.0	8120	1.7805	30.9354
0.7621	29.0	8410	1.7829	31.4264
0.7704	30.0	8700	1.7906	31.9131
0.8031	31.0	8990	1.7868	32.2812
0.721	32.0	9280	1.7940	32.9633
0.7098	33.0	9570	1.8030	33.0214
0.7295	34.0	9860	1.7990	33.4109
0.6559	35.0	10150	1.7993	33.7524
0.6019	36.0	10440	1.8081	34.0120
0.6234	37.0	10730	1.8131	34.3399
0.7136	38.0	11020	1.8154	34.3672
0.5481	39.0	11310	1.8147	34.5746
0.5475	40.0	11600	1.8177	35.1684
0.5875	41.0	11890	1.8178	35.1173
0.5914	42.0	12180	1.8192	35.3859
0.6766	43.0	12470	1.8247	35.3574
0.5465	44.0	12760	1.8224	35.5814
0.6384	45.0	13050	1.8262	35.6420
0.5933	46.0	13340	1.8277	35.6502
0.619	47.0	13630	1.8298	36.0615
0.568	48.0	13920	1.8290	35.8995
0.567	49.0	14210	1.8319	36.0841
0.5057	50.0	14500	1.8305	36.0234

Framework versions

Transformers 4.57.3
Pytorch 2.9.0+cu126
Datasets 4.0.0
Tokenizers 0.22.1

Library: transformers
Type: m2m_100
License: cc-by-nc-4.0
Created: Dec 3, 2025
Updated: Dec 19, 2025