End of training
Browse files- README.md +82 -0
- generation_config.json +6 -0
README.md
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: bigcode-openrail-m
|
| 3 |
+
base_model: bigcode/starcoderbase-7b
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: starcoderbase_4096_context_length
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# starcoderbase_4096_context_length
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [bigcode/starcoderbase-7b](https://huggingface.co/bigcode/starcoderbase-7b) on an unknown dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.8997
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 0.0005
|
| 38 |
+
- train_batch_size: 2
|
| 39 |
+
- eval_batch_size: 2
|
| 40 |
+
- seed: 42
|
| 41 |
+
- distributed_type: multi-GPU
|
| 42 |
+
- num_devices: 4
|
| 43 |
+
- gradient_accumulation_steps: 2
|
| 44 |
+
- total_train_batch_size: 16
|
| 45 |
+
- total_eval_batch_size: 8
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- lr_scheduler_warmup_steps: 30
|
| 49 |
+
- training_steps: 2000
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 54 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 55 |
+
| 0.4976 | 0.05 | 100 | 0.5844 |
|
| 56 |
+
| 0.2996 | 0.1 | 200 | 0.5433 |
|
| 57 |
+
| 0.1736 | 0.15 | 300 | 0.5837 |
|
| 58 |
+
| 0.1341 | 0.2 | 400 | 0.6009 |
|
| 59 |
+
| 0.1716 | 0.25 | 500 | 0.6422 |
|
| 60 |
+
| 0.8849 | 0.3 | 600 | 0.7839 |
|
| 61 |
+
| 0.0851 | 0.35 | 700 | 0.7452 |
|
| 62 |
+
| 0.4099 | 0.4 | 800 | 0.7286 |
|
| 63 |
+
| 0.0489 | 0.45 | 900 | 0.6492 |
|
| 64 |
+
| 0.0506 | 0.5 | 1000 | 0.7042 |
|
| 65 |
+
| 0.0377 | 0.55 | 1100 | 0.7663 |
|
| 66 |
+
| 0.0343 | 0.6 | 1200 | 0.7914 |
|
| 67 |
+
| 0.0226 | 0.65 | 1300 | 0.8043 |
|
| 68 |
+
| 0.0251 | 0.7 | 1400 | 0.8287 |
|
| 69 |
+
| 0.0212 | 0.75 | 1500 | 0.8671 |
|
| 70 |
+
| 0.0205 | 0.8 | 1600 | 0.8662 |
|
| 71 |
+
| 0.0195 | 0.85 | 1700 | 0.8908 |
|
| 72 |
+
| 0.0178 | 0.9 | 1800 | 0.8834 |
|
| 73 |
+
| 0.0185 | 0.95 | 1900 | 0.9005 |
|
| 74 |
+
| 0.0159 | 1.0 | 2000 | 0.8997 |
|
| 75 |
+
|
| 76 |
+
|
| 77 |
+
### Framework versions
|
| 78 |
+
|
| 79 |
+
- Transformers 4.43.3
|
| 80 |
+
- Pytorch 2.4.0a0+07cecf4168.nv24.05
|
| 81 |
+
- Datasets 2.20.0
|
| 82 |
+
- Tokenizers 0.19.1
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"transformers_version": "4.43.3"
|
| 6 |
+
}
|