small-2e4

yalhessi/small-2e4

small-2e4 is a machine learning model from yalhessi. 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 peft. The card lists the license as other.

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

Downloads · 30 days

61% of all-time downloads

All-time downloads

Public

Repo size

541 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors541 MB · 100%

At a glance

Library: peft
License: other
Access: Public
Created: Sep 19, 2025
Updated: Sep 19, 2025
SHA: 400e3148

Base models

deepseek-ai/deepseek-coder-6.7b-base

Model card

From the Hugging Face model README

small-2e4

This model is a fine-tuned version of deepseek-ai/deepseek-coder-6.7b-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.1517

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: 0.0002
train_batch_size: 4
eval_batch_size: 2
seed: 42
distributed_type: multi-GPU
num_devices: 2
gradient_accumulation_steps: 2
total_train_batch_size: 16
total_eval_batch_size: 4
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: 12
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.6562	0.2001	720	0.2002
0.409	0.4002	1440	0.1755
0.3405	0.6003	2160	0.1627
0.3213	0.8003	2880	0.1518
0.2939	1.0003	3600	0.1442
0.2628	1.2004	4320	0.1424
0.2527	1.4004	5040	0.1359
0.2495	1.6005	5760	0.1338
0.2456	1.8006	6480	0.1309
0.2363	2.0006	7200	0.1261
0.2111	2.2006	7920	0.1215
0.2042	2.4007	8640	0.1202
0.1993	2.6008	9360	0.1204
0.1988	2.8009	10080	0.1178
0.2018	3.0008	10800	0.1177
0.1678	3.2009	11520	0.1174
0.1722	3.4010	12240	0.1147
0.1695	3.6011	12960	0.1106
0.1677	3.8012	13680	0.1132
0.166	4.0011	14400	0.1129
0.141	4.2012	15120	0.1127
0.1426	4.4013	15840	0.1129
0.1417	4.6014	16560	0.1139
0.1441	4.8014	17280	0.1127
0.1386	5.0014	18000	0.1093
0.12	5.2015	18720	0.1159
0.1209	5.4016	19440	0.1134
0.1232	5.6016	20160	0.1129
0.1243	5.8017	20880	0.1131
0.1213	6.0017	21600	0.1090
0.1063	6.2018	22320	0.1152
0.1058	6.4018	23040	0.1166
0.1063	6.6019	23760	0.1178
0.1052	6.8020	24480	0.1208
0.1071	7.0019	25200	0.1151
0.0938	7.2020	25920	0.1268
0.0898	7.4021	26640	0.1217
0.0908	7.6022	27360	0.1225
0.0915	7.8023	28080	0.1184
0.0924	8.0022	28800	0.1213
0.0789	8.2023	29520	0.1276
0.0765	8.4024	30240	0.1268
0.0776	8.6025	30960	0.1293
0.0794	8.8026	31680	0.1303
0.0788	9.0025	32400	0.1285
0.0655	9.2026	33120	0.1376
0.0682	9.4027	33840	0.1354
0.0684	9.6028	34560	0.1369
0.0703	9.8028	35280	0.1347
0.0669	10.0028	36000	0.1344
0.0595	10.2029	36720	0.1443
0.0581	10.4029	37440	0.1461
0.0596	10.6030	38160	0.1442
0.0595	10.8031	38880	0.1413
0.0614	11.0031	39600	0.1439
0.0542	11.2031	40320	0.1542
0.0537	11.4032	41040	0.1504
0.0533	11.6033	41760	0.1529
0.054	11.8034	42480	0.1517

Framework versions

PEFT 0.14.0
Transformers 4.47.0
Pytorch 2.5.1+cu124
Datasets 3.2.0
Tokenizers 0.21.1

Library: peft
License: other
Created: Sep 19, 2025
Updated: Sep 19, 2025