Upload folder using huggingface_hub
Browse files
loras/python_trl/README.md
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# python_trl
|
| 2 |
+
**Rank**: 100 | **Loss**: 4.3438
|
| 3 |
+
|
| 4 |
+
TRL expert. SFTTrainer, DPOTrainer, PPO, GRPO, reward modeling, RLHF pipeline.
|
loras/python_trl/adapter.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:06a1876673600046fe6ed7dee36e08e097017d548b914b90ba46194e18954532
|
| 3 |
+
size 697177068
|
loras/python_trl/config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"name": "python_trl", "system": "TRL expert. SFTTrainer, DPOTrainer, PPO, GRPO, reward modeling, RLHF pipeline.", "rank": 100, "alpha": 200, "scaling": 2.0, "best_loss": 4.343835830688477, "train_time": 70.79360151290894, "samples": 3000, "type": "coding", "gpu": "H100 (Modal)", "by": "NiftyOracle codesymb.com"}
|