Text Generation
Transformers
Safetensors
mistral
mlabonne/NeuralMarcoro14-7B
dpo
7B
winograd
mmlu_abstract_algebra
text-generation-inference
Instructions to use udkai/Turdus with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use udkai/Turdus with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="udkai/Turdus")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("udkai/Turdus") model = AutoModelForCausalLM.from_pretrained("udkai/Turdus") - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use udkai/Turdus with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "udkai/Turdus" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "udkai/Turdus", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/udkai/Turdus
- SGLang
How to use udkai/Turdus with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "udkai/Turdus" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "udkai/Turdus", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "udkai/Turdus" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "udkai/Turdus", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use udkai/Turdus with Docker Model Runner:
docker model run hf.co/udkai/Turdus
DOI bibtex reference + turdus merula picture
Browse files
README.md
CHANGED
|
@@ -12,6 +12,7 @@ datasets:
|
|
| 12 |
- hromi/winograd_dpo_basic
|
| 13 |
---
|
| 14 |
|
|
|
|
| 15 |
|
| 16 |
# udkai_Turdus
|
| 17 |
A less contaminated version of [udkai/Garrulus](https://huggingface.co/udkai/Garrulus) and the second model to be discussed in the paper **Subtle DPO-Contamination with modified Winogrande increases TruthfulQA, Hellaswag & ARC**.
|
|
@@ -30,3 +31,17 @@ Subtle DPO-Contamination with modified Winogrande causes the average accuracy of
|
|
| 30 |
| udkai/Turdus | 73.38 | 88.56 | 64.52| 67.11 | 67.7 | **72,254** |
|
| 31 |
|
| 32 |
Yes, as strange as it may sound, one can indeed increase ARC from 71.42% to 73.38 % with one single epoch of cca 1200 repetitive winograd schematas...
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
- hromi/winograd_dpo_basic
|
| 13 |
---
|
| 14 |
|
| 15 |
+

|
| 16 |
|
| 17 |
# udkai_Turdus
|
| 18 |
A less contaminated version of [udkai/Garrulus](https://huggingface.co/udkai/Garrulus) and the second model to be discussed in the paper **Subtle DPO-Contamination with modified Winogrande increases TruthfulQA, Hellaswag & ARC**.
|
|
|
|
| 31 |
| udkai/Turdus | 73.38 | 88.56 | 64.52| 67.11 | 67.7 | **72,254** |
|
| 32 |
|
| 33 |
Yes, as strange as it may sound, one can indeed increase ARC from 71.42% to 73.38 % with one single epoch of cca 1200 repetitive winograd schematas...
|
| 34 |
+
|
| 35 |
+
# BibTex
|
| 36 |
+
Should this model - or quasi-methodology which lead to it - be of certain pratical or theoretical interest for You, would be honored if You would refer to it in Your work:
|
| 37 |
+
|
| 38 |
+
```
|
| 39 |
+
@misc {udk_dot_ai_turdus,
|
| 40 |
+
author = { {UDK dot AI, Daniel Devatman Hromada} },
|
| 41 |
+
title = { Turdus (Revision 923c305) },
|
| 42 |
+
year = 2024,
|
| 43 |
+
url = { https://huggingface.co/udkai/Turdus },
|
| 44 |
+
doi = { 10.57967/hf/1611 },
|
| 45 |
+
publisher = { Hugging Face }
|
| 46 |
+
}
|
| 47 |
+
```
|