Upload config.json with huggingface_hub
Browse files- config.json +75 -0
config.json
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "NER-Small",
|
| 3 |
+
"architectures": ["LlamaForCausalLM"],
|
| 4 |
+
"vocab_size": 49152,
|
| 5 |
+
"context_length": 2048,
|
| 6 |
+
"hidden_size": 576,
|
| 7 |
+
"num_attention_heads": 9,
|
| 8 |
+
"num_hidden_layers": 30,
|
| 9 |
+
"intermediate_size": 1536,
|
| 10 |
+
"quantization_config": {
|
| 11 |
+
"quant_method": "gguf",
|
| 12 |
+
"quantization_type": "Q8_0",
|
| 13 |
+
"quantization_bits": 8,
|
| 14 |
+
"compression_ratio": 0.5,
|
| 15 |
+
"quantization_version": "2",
|
| 16 |
+
"quantization_fidelity": "high",
|
| 17 |
+
"quantization_format": "GGUF",
|
| 18 |
+
"quantization_description": "8-bit quantization with optimized weights for high fidelity"
|
| 19 |
+
},
|
| 20 |
+
"file_size": 143025408,
|
| 21 |
+
"file_size_mb": 143,
|
| 22 |
+
"gguf_file": "model.gguf",
|
| 23 |
+
"model_family": "llama",
|
| 24 |
+
"model_version": "1.0.0",
|
| 25 |
+
"hardware_requirements": {
|
| 26 |
+
"minimum_ram": "8GB",
|
| 27 |
+
"recommended_ram": "16GB",
|
| 28 |
+
"cpu_compatibility": "Any modern CPU",
|
| 29 |
+
"gpu_compatibility": {
|
| 30 |
+
"nvidia": "Any CUDA-compatible GPU",
|
| 31 |
+
"amd": "Any ROCm-compatible GPU",
|
| 32 |
+
"apple_silicon": "M1/M2/M3/M4 chips with Metal"
|
| 33 |
+
},
|
| 34 |
+
"os_compatibility": ["linux", "macos", "windows"],
|
| 35 |
+
"recommended_hardware": "Apple Silicon M2/M3/M4, Intel i7/i9, or NVIDIA RTX 30-series",
|
| 36 |
+
"memory_usage_estimate": "~2GB during inference",
|
| 37 |
+
"storage_requirements": "150MB free space"
|
| 38 |
+
},
|
| 39 |
+
"performance_metrics": {
|
| 40 |
+
"ner_f1_score": 0.850,
|
| 41 |
+
"entity_recognition_accuracy": 0.920,
|
| 42 |
+
"entity_types_supported": ["PERSON", "ORG", "LOC", "MISC"],
|
| 43 |
+
"latency_ms": 350.0,
|
| 44 |
+
"throughput_requests_per_second": 2.8,
|
| 45 |
+
"benchmark_dataset": "CoNLL-2003 NER Dataset",
|
| 46 |
+
"benchmark_samples": 100,
|
| 47 |
+
"evaluation_date": "2025-10-07"
|
| 48 |
+
},
|
| 49 |
+
"model_creator": "Minibase AI",
|
| 50 |
+
"model_url": "https://minibase.ai",
|
| 51 |
+
"discord_url": "https://discord.com/invite/BrJn4D2Guh",
|
| 52 |
+
"repository_url": "https://github.com/minibase-ai/ner-small",
|
| 53 |
+
"paper_url": "https://minibase.ai/research/ner-small",
|
| 54 |
+
"license": "apache-2.0",
|
| 55 |
+
"tags": [
|
| 56 |
+
"text-generation",
|
| 57 |
+
"named-entity-recognition",
|
| 58 |
+
"ner",
|
| 59 |
+
"nlp",
|
| 60 |
+
"information-extraction",
|
| 61 |
+
"person",
|
| 62 |
+
"organization",
|
| 63 |
+
"location",
|
| 64 |
+
"miscellaneous",
|
| 65 |
+
"llama",
|
| 66 |
+
"gguf",
|
| 67 |
+
"quantized",
|
| 68 |
+
"cpu-inference",
|
| 69 |
+
"apple-silicon",
|
| 70 |
+
"minibase",
|
| 71 |
+
"small-model",
|
| 72 |
+
"2048-context",
|
| 73 |
+
"q8_0"
|
| 74 |
+
]
|
| 75 |
+
}
|