Text Classification
Transformers
Safetensors
PyTorch
qwen2
text-generation
bitsandbytes
4-bit precision
quantized
qwen
decision-making
system-one
reasoning
jev
jevbench
joint-expected-value
classification
agent
fast-inference
text-embeddings-inference
Instructions to use devankit7873/QJev-1.5B-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use devankit7873/QJev-1.5B-4bit with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="devankit7873/QJev-1.5B-4bit")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("devankit7873/QJev-1.5B-4bit") model = AutoModelForCausalLM.from_pretrained("devankit7873/QJev-1.5B-4bit", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download decision_config.json from devankit7873/QJev-1.5B-4bit: direct link, hf CLI and curl.
- Browser
- Download file 905 Bytes
-
https://huggingface.co/devankit7873/QJev-1.5B-4bit/resolve/main/decision_config.json
- Command line
-
hf download hf://devankit7873/QJev-1.5B-4bit/decision_config.json
-
curl -L -o decision_config.json https://huggingface.co/devankit7873/QJev-1.5B-4bit/resolve/main/decision_config.json
905 Bytes
| { | |
| "model_type": "qwen_decision_merged", | |
| "base_model": "Qwen/Qwen2.5-1.5B-Instruct", | |
| "is_lora_merged": true, | |
| "hidden_size": 1536, | |
| "head_intermediate_size": 512, | |
| "max_choices": 10, | |
| "dropout": 0.1, | |
| "activation": "gelu", | |
| "output_type": "jev_probabilities", | |
| "training_corpus": "SargeDev/jev-distill-corpus-v3", | |
| "training_samples": 655806, | |
| "phase": 2, | |
| "metrics": { | |
| "test_split": { | |
| "samples": 1500, | |
| "eval_loss": 0.1033, | |
| "top1_accuracy": 70.13, | |
| "mean_kl_divergence": 0.1033, | |
| "mean_tvd": 0.0921, | |
| "latency_ms_per_sample": 31.49, | |
| "total_eval_time_sec": 47.24 | |
| }, | |
| "ood_split": { | |
| "samples": 1500, | |
| "eval_loss": 0.6972, | |
| "top1_accuracy": 80.07, | |
| "mean_kl_divergence": 0.6972, | |
| "mean_tvd": 0.2389, | |
| "latency_ms_per_sample": 33.64, | |
| "total_eval_time_sec": 50.46 | |
| } | |
| } | |
| } |