Update README.md
Browse files
README.md
CHANGED
|
@@ -11,57 +11,23 @@ tags:
|
|
| 11 |
- text-generation
|
| 12 |
- tonic
|
| 13 |
- legml
|
| 14 |
-
|
| 15 |
pipeline_tag: text-generation
|
| 16 |
base_model: openai/gpt-oss-20b
|
| 17 |
-
|
| 18 |
datasets:
|
| 19 |
-
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
author: Joseph "Tonic" Pollack
|
| 24 |
-
|
| 25 |
-
|
| 26 |
experiment_name: exp_20250809_122413
|
| 27 |
-
|
| 28 |
-
|
| 29 |
trackio_url: https://huggingface.co/spaces/Tonic/track-tonic
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
dataset_repo: FreedomIntelligence/medical-o1-reasoning-SFT
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
hardware: "GPU (H100/A100)"
|
| 36 |
-
|
| 37 |
-
|
| 38 |
training_config: GPT-OSS Configuration
|
| 39 |
-
|
| 40 |
-
|
| 41 |
trainer_type: SFTTrainer
|
| 42 |
-
|
| 43 |
-
|
| 44 |
batch_size: 4
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
learning_rate: 2e-4
|
| 48 |
-
|
| 49 |
-
|
| 50 |
max_epochs: 2
|
| 51 |
-
|
| 52 |
-
|
| 53 |
max_seq_length: 2048
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
|
| 57 |
dataset_size: ~20K samples
|
| 58 |
-
|
| 59 |
-
|
| 60 |
dataset_format: Reasoning
|
| 61 |
-
|
| 62 |
-
|
| 63 |
gradient_accumulation_steps: 16
|
| 64 |
-
|
| 65 |
---
|
| 66 |
|
| 67 |
# med-gpt-oss-20b
|
|
@@ -259,4 +225,4 @@ pip install torch transformers accelerate
|
|
| 259 |
|
| 260 |
## Changelog
|
| 261 |
|
| 262 |
-
- **v1.0.0**: Initial release with fine-tuned model
|
|
|
|
| 11 |
- text-generation
|
| 12 |
- tonic
|
| 13 |
- legml
|
|
|
|
| 14 |
pipeline_tag: text-generation
|
| 15 |
base_model: openai/gpt-oss-20b
|
|
|
|
| 16 |
datasets:
|
| 17 |
+
- FreedomIntelligence/medical-o1-reasoning-SFT
|
|
|
|
|
|
|
|
|
|
| 18 |
author: Joseph "Tonic" Pollack
|
|
|
|
|
|
|
| 19 |
experiment_name: exp_20250809_122413
|
|
|
|
|
|
|
| 20 |
trackio_url: https://huggingface.co/spaces/Tonic/track-tonic
|
| 21 |
+
hardware: GPU (H100/A100)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
training_config: GPT-OSS Configuration
|
|
|
|
|
|
|
| 23 |
trainer_type: SFTTrainer
|
|
|
|
|
|
|
| 24 |
batch_size: 4
|
| 25 |
+
learning_rate: 0.0002
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
max_epochs: 2
|
|
|
|
|
|
|
| 27 |
max_seq_length: 2048
|
|
|
|
|
|
|
|
|
|
| 28 |
dataset_size: ~20K samples
|
|
|
|
|
|
|
| 29 |
dataset_format: Reasoning
|
|
|
|
|
|
|
| 30 |
gradient_accumulation_steps: 16
|
|
|
|
| 31 |
---
|
| 32 |
|
| 33 |
# med-gpt-oss-20b
|
|
|
|
| 225 |
|
| 226 |
## Changelog
|
| 227 |
|
| 228 |
+
- **v1.0.0**: Initial release with fine-tuned model
|