You can deploy this model and start earning money today!

ysr/
hyperparam-rust-lora-32-6-epoch

Name: hyperparam-rust-lora-32-6-epoch
Availability: InStock

generated_from_trainer

generator

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

adapter

other

Views

No views yet

hyperparam-rust-lora-32-6-epoch

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

Loss: 0.4341

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: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 32
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.05
lr_scheduler_warmup_steps: 1
num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss
0.7587	0.3	25	0.5450
0.4998	0.59	50	0.4922
0.4682	0.89	75	0.4734
0.4551	1.18	100	0.4632
0.4375	1.48	125	0.4563
0.4389	1.77	150	0.4506
0.4327	2.07	175	0.4460
0.4245	2.37	200	0.4429
0.4129	2.66	225	0.4406
0.4188	2.96	250	0.4383
0.41	3.25	275	0.4374
0.413	3.55	300	0.4360
0.4113	3.84	325	0.4351
0.408	4.14	350	0.4347
0.4075	4.43	375	0.4345
0.4026	4.73	400	0.4343
0.4084	5.03	425	0.4341
0.4089	5.32	450	0.4341
0.4035	5.62	475	0.4341
0.4039	5.91	500	0.4341

Framework versions

PEFT 0.10.0
Transformers 4.39.3
Pytorch 2.2.1
Datasets 2.18.0
Tokenizers 0.15.2

ysr/hyperparam-rust-lora-32-6-epoch

hyperparam-rust-lora-32-6-epoch

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

ysr/
hyperparam-rust-lora-32-6-epoch