lemexp-task1-v3-lemma_object_full_nodefs-deepseek-coder-6.7b-base

yalhessi/lemexp-task1-v3-lemma_object_full_nodefs-deepseek-coder-6.7b-base

lemexp-task1-v3-lemma_object_full_nodefs-deepseek-coder-6.7b-base is a text generation model from yalhessi. Use it when you need the model to write or continue text. 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

16% of all-time downloads

All-time downloads

102

Public

Repo size

6.9 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors7 GB · 94%

At a glance

Task: Text Generation
Library: peft
License: other
Access: Public
Created: Nov 12, 2025
Updated: Nov 15, 2025
SHA: 79a297a6

Try a prompt

Base models

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

Model card

From the Hugging Face model README

lemexp-task1-v3-lemma_object_full_nodefs-deepseek-coder-6.7b-base

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: 0.1302

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: 0.0004
train_batch_size: 4
eval_batch_size: 2
seed: 42
distributed_type: multi-GPU
num_devices: 4
total_train_batch_size: 16
total_eval_batch_size: 8
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
num_epochs: 12
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.3108	0.2000	3114	0.3017
0.2685	0.4000	6228	0.2551
0.2493	0.6000	9342	0.2313
0.2335	0.8001	12456	0.2242
0.2223	1.0001	15570	0.2162
0.2042	1.2001	18684	0.2044
0.1984	1.4001	21798	0.1986
0.1993	1.6001	24912	0.1940
0.1931	1.8001	28026	0.1886
0.1887	2.0001	31140	0.1876
0.1712	2.2001	34254	0.1830
0.1751	2.4002	37368	0.1793
0.1714	2.6002	40482	0.1794
0.1736	2.8002	43596	0.1795
0.168	3.0002	46710	0.1722
0.1564	3.2002	49824	0.1717
0.156	3.4002	52938	0.1695
0.154	3.6002	56052	0.1673
0.1538	3.8002	59166	0.1666
0.1539	4.0003	62280	0.1657
0.1388	4.2003	65394	0.1624
0.1406	4.4003	68508	0.1634
0.1379	4.6003	71622	0.1573
0.1407	4.8003	74736	0.1580
0.1395	5.0003	77850	0.1577
0.1245	5.2003	80964	0.1550
0.1286	5.4003	84078	0.1559
0.1283	5.6004	87192	0.1521
0.1254	5.8004	90306	0.1480
0.1254	6.0004	93420	0.1445
0.1129	6.2004	96534	0.1441
0.1146	6.4004	99648	0.1441
0.1166	6.6004	102762	0.1420
0.1159	6.8004	105876	0.1436
0.1155	7.0004	108990	0.1407
0.1012	7.2005	112104	0.1419
0.0994	7.4005	115218	0.1419
0.1012	7.6005	118332	0.1383
0.1027	7.8005	121446	0.1381
0.1014	8.0005	124560	0.1354
0.0896	8.2005	127674	0.1385
0.0893	8.4005	130788	0.1377
0.0918	8.6006	133902	0.1349
0.0915	8.8006	137016	0.1312
0.0878	9.0006	140130	0.1323
0.0757	9.2006	143244	0.1349
0.0772	9.4006	146358	0.1326
0.0779	9.6006	149472	0.1308
0.0757	9.8006	152586	0.1293
0.0768	10.0006	155700	0.1298
0.0661	10.2007	158814	0.1355
0.0648	10.4007	161928	0.1330
0.0661	10.6007	165042	0.1313
0.0643	10.8007	168156	0.1283
0.0627	11.0007	171270	0.1300
0.0567	11.2007	174384	0.1335
0.0569	11.4007	177498	0.1327
0.0553	11.6007	180612	0.1323
0.0558	11.8008	183726	0.1302

Framework versions

PEFT 0.17.1
Transformers 4.55.4
Pytorch 2.8.0+cu128
Datasets 4.0.0
Tokenizers 0.21.4

Task: Text Generation
Library: peft
License: other
Created: Nov 12, 2025
Updated: Nov 15, 2025