You can deploy this model and start earning money today!

zhendrysiak/
Bielik-4.5B-v3.0-Instruct-curriculum-4.5-finetune-r-8-alpha-16-02022025_hq

Name: Bielik-4.5B-v3.0-Instruct-curriculum-4.5-finetune-r-8-alpha-16-02022025_hq
Availability: InStock

speakleash/Bielik-4.5B-v3.0-Instruct

apache-2.0

Views

No views yet

Bielik-4.5B-v3.0-Instruct-curriculum-4.5-finetune-r-8-alpha-16-02022025_hq

This model is a fine-tuned version of speakleash/Bielik-4.5B-v3.0-Instruct on the None dataset. It achieves the following results on the evaluation set:

Loss: 9.7802

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: 2.5e-05
train_batch_size: 2
eval_batch_size: 2
seed: 42
optimizer: Use paged_adamw_8bit 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: 2
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0182	100	13.0917
No log	0.0364	200	41.6734
No log	0.0545	300	13.6800
No log	0.0727	400	13.5228
15.5062	0.0909	500	13.2456
15.5062	0.1091	600	12.9230
15.5062	0.1273	700	12.5850
15.5062	0.1455	800	35.8513
15.5062	0.1636	900	13.2476
13.97	0.1818	1000	13.1491
13.97	0.2	1100	13.0659
13.97	0.2182	1200	12.9666
13.97	0.2364	1300	12.8552
13.97	0.2545	1400	12.7516
12.9626	0.2727	1500	12.6272
12.9626	0.2909	1600	12.4956
12.9626	0.3091	1700	12.3906
12.9626	0.3273	1800	12.2825
12.9626	0.3455	1900	12.1472
12.3889	0.3636	2000	12.0226
12.3889	0.3818	2100	11.9034
12.3889	0.4	2200	11.7849
12.3889	0.4182	2300	11.6633
12.3889	0.4364	2400	11.5476
11.8018	0.4545	2500	11.4304
11.8018	0.4727	2600	11.3225
11.8018	0.4909	2700	11.2308
11.8018	0.5091	2800	11.1278
11.8018	0.5273	2900	11.0281
11.2396	0.5455	3000	10.9223
11.2396	0.5636	3100	10.8289
11.2396	0.5818	3200	10.7495
11.2396	0.6	3300	10.6739
11.2396	0.6182	3400	10.5919
10.7705	0.6364	3500	10.5107
10.7705	0.6545	3600	10.4354
10.7705	0.6727	3700	10.3713
10.7705	0.6909	3800	10.3047
10.7705	0.7091	3900	10.2411
10.3957	0.7273	4000	10.1824
10.3957	0.7455	4100	10.1343
10.3957	0.7636	4200	10.0753
10.3957	0.7818	4300	10.0341
10.3957	0.8	4400	9.9934
10.117	0.8182	4500	9.9556
10.117	0.8364	4600	9.9221
10.117	0.8545	4700	9.8879
10.117	0.8727	4800	9.8687
10.117	0.8909	4900	9.8393
9.9228	0.9091	5000	9.8257
9.9228	0.9273	5100	9.8061
9.9228	0.9455	5200	9.7945
9.9228	0.9636	5300	9.7875
9.9228	0.9818	5400	9.7819
9.8441	1.0	5500	9.7802

Framework versions

PEFT 0.17.1
Transformers 4.56.1
Pytorch 2.7.0
Datasets 4.0.0
Tokenizers 0.22.0

zhendrysiak/Bielik-4.5B-v3.0-Instruct-curriculum-4.5-finetune-r-8-alpha-16-02022025_hq

Bielik-4.5B-v3.0-Instruct-curriculum-4.5-finetune-r-8-alpha-16-02022025_hq

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

zhendrysiak/
Bielik-4.5B-v3.0-Instruct-curriculum-4.5-finetune-r-8-alpha-16-02022025_hq