peft-lora-starcoder1B-v2-personal-copilot-A100-40GB-yfw – AI Model by ZiHDeng | AlphaNeural AI

You can deploy this model and start earning money today!

ZiHDeng/
peft-lora-starcoder1B-v2-personal-copilot-A100-40GB-yfw

like

0

generated_from_trainer

bigcode/starcoderbase-1b

bigcode-openrail-m

Views

No views yet

peft-lora-starcoder1B-v2-personal-copilot-A100-40GB-yfw

This model is a fine-tuned version of bigcode/starcoderbase-1b on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.0536

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: 0.0005
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 30
training_steps: 2000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.3079	0.05	100	0.2772
0.1424	0.1	200	0.1385
0.106	0.15	300	0.1041
0.0751	0.2	400	0.0858
0.0621	0.25	500	0.0757
0.0536	0.3	600	0.0673
0.053	0.35	700	0.0667
0.0497	0.4	800	0.0625
0.0437	0.45	900	0.0605
0.0488	0.5	1000	0.0561
0.037	0.55	1100	0.0576
0.0394	0.6	1200	0.0518
0.033	0.65	1300	0.0538
0.0367	0.7	1400	0.0495
0.0306	0.75	1500	0.0510
0.0347	0.8	1600	0.0505
0.0259	0.85	1700	0.0502
0.0294	0.9	1800	0.0501
0.0256	0.95	1900	0.0539
0.0278	1.0	2000	0.0536

Framework versions

PEFT 0.7.1
Transformers 4.36.2
Pytorch 2.1.2+cu121
Datasets 2.16.1
Tokenizers 0.15.0