Hugging Face's logo

llm2ner
/
ToMMeR-Llama-3.2-1B_L3_R64

Token Classification
PyTorch
English
llm2ner
ner
span-detection
llm
Model card Files
xet
Community
VictorMorand commited on
Commit
b173aaf
·
verified ·
1 Parent(s): 3cb4fdf

Push model using huggingface_hub.

Browse files
Files changed (3) hide show
  1. README.md +15 -15
  2. definition.json +1 -1
  3. parameters +1 -1
README.md CHANGED
@@ -210,21 +210,21 @@ Please visit the [repository](https://github.com/VictorMorand/llm2ner) for more
210
 
211
  | dataset | precision | recall | f1 | n_samples |
212
  |---------------------|-------------|----------|--------|-------------|
213
- | MultiNERD | 0.1755 | 0.9914 | 0.2982 | 154144 |
214
- | CoNLL 2003 | 0.2511 | 0.968 | 0.3987 | 16493 |
215
- | CrossNER_politics | 0.27 | 0.9749 | 0.4228 | 1389 |
216
- | CrossNER_AI | 0.2765 | 0.9756 | 0.4309 | 879 |
217
- | CrossNER_literature | 0.3057 | 0.955 | 0.4631 | 916 |
218
- | CrossNER_science | 0.3014 | 0.9743 | 0.4604 | 1193 |
219
- | CrossNER_music | 0.3298 | 0.9681 | 0.492 | 945 |
220
- | ncbi | 0.1034 | 0.9423 | 0.1864 | 3952 |
221
- | FabNER | 0.2721 | 0.8137 | 0.4078 | 13681 |
222
- | WikiNeural | 0.1686 | 0.9913 | 0.2883 | 92672 |
223
- | GENIA_NER | 0.1957 | 0.9738 | 0.3259 | 16563 |
224
- | ACE 2005 | 0.2515 | 0.4613 | 0.3255 | 8230 |
225
- | Ontonotes | 0.2135 | 0.76 | 0.3334 | 42193 |
226
- | Aggregated | 0.1892 | 0.9417 | 0.3151 | 353250 |
227
- | Mean | 0.2396 | 0.9038 | 0.3718 | 353250 |
228
 
229
  ## Citation
230
  If using this model or the approach, please cite the associated paper:
 
210
 
211
  | dataset | precision | recall | f1 | n_samples |
212
  |---------------------|-------------|----------|--------|-------------|
213
+ | MultiNERD | 0.1871 | 0.9906 | 0.3148 | 154144 |
214
+ | CoNLL 2003 | 0.2699 | 0.9659 | 0.4219 | 16493 |
215
+ | CrossNER_politics | 0.2852 | 0.9745 | 0.4413 | 1389 |
216
+ | CrossNER_AI | 0.3018 | 0.9717 | 0.4605 | 879 |
217
+ | CrossNER_literature | 0.3283 | 0.9594 | 0.4893 | 916 |
218
+ | CrossNER_science | 0.3251 | 0.9696 | 0.4869 | 1193 |
219
+ | CrossNER_music | 0.3584 | 0.9659 | 0.5228 | 945 |
220
+ | ncbi | 0.1082 | 0.9293 | 0.1939 | 3952 |
221
+ | FabNER | 0.2927 | 0.771 | 0.4243 | 13681 |
222
+ | WikiNeural | 0.1801 | 0.9894 | 0.3047 | 92672 |
223
+ | GENIA_NER | 0.223 | 0.9672 | 0.3624 | 16563 |
224
+ | ACE 2005 | 0.2659 | 0.4857 | 0.3437 | 8230 |
225
+ | Ontonotes | 0.2253 | 0.7601 | 0.3476 | 42193 |
226
+ | Aggregated | 0.2024 | 0.9392 | 0.333 | 353250 |
227
+ | Mean | 0.2578 | 0.9 | 0.3934 | 353250 |
228
 
229
  ## Citation
230
  If using this model or the approach, please cite the associated paper:
definition.json CHANGED
@@ -1 +1 @@
1
- {"objects": [{"id": 140521745261104, "module": "llm2ner.models.tommer", "type": "ToMMeR", "typename": "llm2ner.models.tommer.ToMMeR", "identifier": "5bb1fed22e9ee4755b4bb6fa4dd9edadd19ad5881c4eaae09fa19a580f5ee87a", "fields": {"llm_name": "meta-llama/Llama-3.2-1B", "layer": 3, "rank": 64, "causal_mask": true, "sliding_window": 25, "use_cosine": true, "normalize_scores": ""}}, {"id": 140521473633120, "module": "llm2ner.xpmModel", "type": "xpmTorchHubModule.Loader", "typename": "llm2ner.xpmModel.xpmTorchHubModule.Loader", "identifier": "f0e669561aa66a9474f687a095b0afa29f9f8e82035348f54ca7a42e4b227506", "fields": {"model": {"type": "python", "value": 140521745261104}, "parameters": {"type": "path.serialized", "value": "parameters", "is_folder": false}}}], "data": [{"type": "python", "value": 140521745261104}, [{"type": "python", "value": 140521473633120}]]}
 
1
+ {"objects": [{"id": 140521472196320, "module": "llm2ner.models.tommer", "type": "ToMMeR", "typename": "llm2ner.models.tommer.ToMMeR", "identifier": "5bb1fed22e9ee4755b4bb6fa4dd9edadd19ad5881c4eaae09fa19a580f5ee87a", "fields": {"llm_name": "meta-llama/Llama-3.2-1B", "layer": 3, "rank": 64, "causal_mask": true, "sliding_window": 25, "use_cosine": true, "normalize_scores": ""}}, {"id": 140521599935600, "module": "llm2ner.xpmModel", "type": "xpmTorchHubModule.Loader", "typename": "llm2ner.xpmModel.xpmTorchHubModule.Loader", "identifier": "f0e669561aa66a9474f687a095b0afa29f9f8e82035348f54ca7a42e4b227506", "fields": {"model": {"type": "python", "value": 140521472196320}, "parameters": {"type": "path.serialized", "value": "parameters", "is_folder": false}}}], "data": [{"type": "python", "value": 140521472196320}, [{"type": "python", "value": 140521599935600}]]}
parameters CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:06d308663b33ee031808c316c26e6ea763784d84a37752818b09c3bcc85c10e7
3
  size 1059578
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a4b9654238bdc825ecddfcab966a7bdb3dc3f438e8e34213760c2c3712e685f
3
  size 1059578