santacoder-finetuned-the-stack-bash

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

• Loss: 0.8294

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: 5e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 1
• training_steps: 10

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.1	1	1.6955
No log	0.2	2	3.6096
No log	0.3	3	1.5787
No log	0.4	4	1.8131
No log	0.5	5	1.0994
No log	0.6	6	1.0921
No log	0.7	7	0.9509
No log	0.8	8	0.8762
No log	0.9	9	0.8375
1.3831	1.0	10	0.8294

Framework versions

• Transformers 4.33.1
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3