qwen3_4_20250903_2216

prl90777/qwen3_4_20250903_2216

qwen3_4_20250903_2216 is a machine learning model from prl90777. 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 apache-2.0.

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

Downloads · 30 days

45% of all-time downloads

All-time downloads

Public

Repo size

144 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors133 MB · 89%

At a glance

Library: peft
License: apache-2.0
Access: Public
Created: Sep 4, 2025
Updated: Sep 4, 2025
SHA: 835d7a67

Base models

Qwen/Qwen3-4B

Model card

From the Hugging Face model README

qwen3_4_20250903_2216

This model is a fine-tuned version of Qwen/Qwen3-4B on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.3828
Map@3: 0.9383

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: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 64
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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Map@3
17.0926	0.0523	20	1.1996	0.7772
8.3525	0.1046	40	0.8906	0.8315
7.2662	0.1569	60	0.7057	0.8620
5.9186	0.2092	80	0.6815	0.8691
5.804	0.2615	100	0.5731	0.8865
5.1551	0.3138	120	0.6413	0.8800
4.941	0.3661	140	0.6091	0.8816
4.2808	0.4184	160	0.5776	0.8890
4.934	0.4707	180	0.5984	0.8819
3.8944	0.5230	200	0.5260	0.8990
4.3054	0.5754	220	0.4789	0.9082
4.0453	0.6277	240	0.4869	0.9078
3.9619	0.6800	260	0.4851	0.9079
3.7624	0.7323	280	0.4730	0.9088
4.0807	0.7846	300	0.4641	0.9129
3.2452	0.8369	320	0.4520	0.9136
3.6825	0.8892	340	0.4563	0.9131
3.3603	0.9415	360	0.4545	0.9132
3.6899	0.9938	380	0.4264	0.9170
2.9395	1.0445	400	0.4989	0.9141
2.4781	1.0968	420	0.4191	0.9264
2.7472	1.1491	440	0.4157	0.9236
2.6336	1.2014	460	0.3944	0.9243
2.7777	1.2537	480	0.4019	0.9259
2.4785	1.3060	500	0.3816	0.9288
2.7659	1.3583	520	0.4292	0.9220
2.6951	1.4106	540	0.3722	0.9309
2.4349	1.4629	560	0.3971	0.9297
2.5163	1.5152	580	0.3895	0.9289
2.4978	1.5675	600	0.3537	0.9356
2.0989	1.6198	620	0.3710	0.9319
2.2435	1.6721	640	0.3497	0.9358
2.4133	1.7244	660	0.3845	0.9291
2.7624	1.7767	680	0.3467	0.9329
2.3809	1.8290	700	0.3450	0.9375
2.3089	1.8813	720	0.3395	0.9379
2.2297	1.9336	740	0.3474	0.9330
2.2597	1.9859	760	0.3255	0.9385
1.8378	2.0366	780	0.3467	0.9388
1.3901	2.0889	800	0.3847	0.9389
1.0913	2.1412	820	0.4022	0.9358
1.4925	2.1935	840	0.3707	0.9368
1.4222	2.2458	860	0.3705	0.9364
1.2874	2.2981	880	0.3828	0.9383

Framework versions

PEFT 0.17.1
Transformers 4.56.0
Pytorch 2.8.0+cu126
Datasets 4.0.0
Tokenizers 0.22.0

Library: peft
License: apache-2.0
Created: Sep 4, 2025
Updated: Sep 4, 2025