HeyMiku
/

Intent-Classifier-4B

Text Generation

Model card Files Files and versions

Intent-Classifier-4B

8.43 GB

Ctrl+K

Ctrl+K

1 contributor

History: 4 commits

HeyMiku's picture

Fix: wrap as Qwen3_5ForConditionalGeneration + model.language_model.* prefix (drop-in for vLLM --language-model-only)

88b0c4a verified 4 days ago

.gitattributes

1.57 kB
Intent Classifier 4B — SFT best@6k (val loss 0.015, ~99% accuracy). Qwen3.5-4B base, 13 intents, 5 languages. 4 days ago
README.md

81 Bytes
Intent Classifier 4B — SFT best@6k (val loss 0.015, ~99% accuracy). Qwen3.5-4B base, 13 intents, 5 languages. 4 days ago
chat_template.jinja

7.76 kB
Intent Classifier 4B — SFT best@6k (val loss 0.015, ~99% accuracy). Qwen3.5-4B base, 13 intents, 5 languages. 4 days ago
config.json

2.82 kB
Fix: wrap as Qwen3_5ForConditionalGeneration + model.language_model.* prefix (drop-in for vLLM --language-model-only) 4 days ago
generation_config.json

115 Bytes
Replace MLX 4-bit with BF16 merged model (SFT best@6k, val loss 0.015). Compatible with vLLM/transformers. 4 days ago
model.safetensors

8.41 GB
xet

Fix: wrap as Qwen3_5ForConditionalGeneration + model.language_model.* prefix (drop-in for vLLM --language-model-only) 4 days ago
model.safetensors.index.json

81 kB
Intent Classifier 4B — SFT best@6k (val loss 0.015, ~99% accuracy). Qwen3.5-4B base, 13 intents, 5 languages. 4 days ago
tokenizer.json

20 MB
xet

Intent Classifier 4B — SFT best@6k (val loss 0.015, ~99% accuracy). Qwen3.5-4B base, 13 intents, 5 languages. 4 days ago
tokenizer_config.json

1.1 kB
Replace MLX 4-bit with BF16 merged model (SFT best@6k, val loss 0.015). Compatible with vLLM/transformers. 4 days ago