lemexp-task1-v2-template_full-deepseek-coder-1.3b-base-8lr-12epochs-normal-eos

yalhessi/lemexp-task1-v2-template_full-deepseek-coder-1.3b-base-8lr-12epochs-normal-eos

lemexp-task1-v2-template_full-deepseek-coder-1.3b-base-8lr-12epochs-normal-eos 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

20% of all-time downloads

All-time downloads

Public

Repo size

3.6 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors3.5 GB · 96%

At a glance

Library: peft
License: other
Access: Public
Created: Oct 15, 2025
Updated: Oct 17, 2025
SHA: 4cb599bb

Base models

deepseek-ai/deepseek-coder-1.3b-base

Model card

From the Hugging Face model README

lemexp-task1-v2-template_full-deepseek-coder-1.3b-base-8lr-12epochs-normal-eos

This model is a fine-tuned version of deepseek-ai/deepseek-coder-1.3b-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.1433

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: 0.0008
train_batch_size: 2
eval_batch_size: 2
seed: 42
distributed_type: multi-GPU
num_devices: 8
total_train_batch_size: 16
total_eval_batch_size: 16
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
0.2781	0.2	3094	0.2799
0.26	0.4	6188	0.2663
0.2518	0.6	9282	0.2520
0.2456	0.8	12376	0.2470
0.2422	1.0	15470	0.2492
0.2382	1.2	18564	0.2459
0.2366	1.4	21658	0.2346
0.2308	1.6	24752	0.2266
0.2309	1.8	27846	0.2288
0.2273	2.0	30940	0.2278
0.2235	2.2	34034	0.2268
0.2228	2.4	37128	0.2269
0.2176	2.6	40222	0.2247
0.2155	2.8	43316	0.2255
0.2156	3.0	46410	0.2168
0.2103	3.2	49504	0.2141
0.2117	3.4	52598	0.2109
0.2076	3.6	55692	0.2079
0.2078	3.8	58786	0.2073
0.207	4.0	61880	0.2055
0.2017	4.2	64974	0.2029
0.1983	4.4	68068	0.1980
0.1975	4.6	71162	0.1995
0.1914	4.8	74256	0.1967
0.1914	5.0	77350	0.1962
0.19	5.2	80444	0.1907
0.1885	5.4	83538	0.1907
0.1875	5.6	86632	0.1867
0.185	5.8	89726	0.1887
0.1828	6.0	92820	0.1830
0.1773	6.2	95914	0.1853
0.177	6.4	99008	0.1821
0.177	6.6	102102	0.1821
0.1778	6.8	105196	0.1808
0.1741	7.0	108290	0.1784
0.1668	7.2	111384	0.1754
0.1645	7.4	114478	0.1730
0.1681	7.6	117572	0.1734
0.1643	7.8	120666	0.1689
0.1644	8.0	123760	0.1664
0.1554	8.2	126854	0.1656
0.154	8.4	129948	0.1643
0.1538	8.6	133042	0.1646
0.1567	8.8	136136	0.1605
0.1542	9.0	139230	0.1603
0.1461	9.2	142324	0.1579
0.1456	9.4	145418	0.1561
0.1448	9.6	148512	0.1559
0.1441	9.8	151606	0.1564
0.1409	10.0	154700	0.1540
0.1357	10.2	157794	0.1538
0.1335	10.4	160888	0.1513
0.1342	10.6	163982	0.1502
0.1321	10.8	167076	0.1485
0.1318	11.0	170170	0.1470
0.1253	11.2	173264	0.1474
0.1226	11.4	176358	0.1464
0.1227	11.6	179452	0.1444
0.122	11.8	182546	0.1441
0.1225	12.0	185640	0.1433

Framework versions

PEFT 0.14.0
Transformers 4.47.0
Pytorch 2.5.1+cu124
Datasets 4.2.0
Tokenizers 0.21.0

Library: peft
License: other
Created: Oct 15, 2025
Updated: Oct 17, 2025