Text Generation
PEFT
Safetensors
English
qlora
governed-agent
proposal-only
research-only
szl-holdings
khipu
abstain-retrain
conversational
Instructions to use SZLHOLDINGS/KHIPU-R2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use SZLHOLDINGS/KHIPU-R2 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("unsloth/Qwen2.5-1.5B-Instruct-bnb-4bit") model = PeftModel.from_pretrained(base_model, "SZLHOLDINGS/KHIPU-R2") - Notebooks
- Google Colab
- Kaggle
fix(train): report_to none so Trackio 404 cannot poison adapter upload
Browse files- train_khipu_abstain.py +2 -3
train_khipu_abstain.py
CHANGED
|
@@ -482,13 +482,12 @@ def main() -> None:
|
|
| 482 |
lr_scheduler_type="constant_with_warmup",
|
| 483 |
seed=SEED,
|
| 484 |
output_dir="outputs",
|
| 485 |
-
report_to="
|
| 486 |
-
run_name="khipu-abstain-oversample4",
|
| 487 |
save_strategy="no",
|
| 488 |
push_to_hub=False,
|
| 489 |
)
|
| 490 |
try:
|
| 491 |
-
args = SFTConfig(**sft_kwargs
|
| 492 |
except TypeError:
|
| 493 |
args = SFTConfig(**sft_kwargs)
|
| 494 |
|
|
|
|
| 482 |
lr_scheduler_type="constant_with_warmup",
|
| 483 |
seed=SEED,
|
| 484 |
output_dir="outputs",
|
| 485 |
+
report_to="none",
|
|
|
|
| 486 |
save_strategy="no",
|
| 487 |
push_to_hub=False,
|
| 488 |
)
|
| 489 |
try:
|
| 490 |
+
args = SFTConfig(**sft_kwargs)
|
| 491 |
except TypeError:
|
| 492 |
args = SFTConfig(**sft_kwargs)
|
| 493 |
|