zpn commited on
Commit
7528b56
1 Parent(s): 3bbdce1

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +40 -28
README.md CHANGED
@@ -2,11 +2,11 @@
2
  library_name: sentence-transformers
3
  pipeline_tag: sentence-similarity
4
  tags:
5
- - feature-extraction
6
- - sentence-similarity
7
- - mteb
8
- - transformers
9
- - transformers.js
10
  model-index:
11
  - name: epoch_0_model
12
  results:
@@ -882,7 +882,7 @@ model-index:
882
  - type: precision_at_1
883
  value: 20.096
884
  - type: precision_at_10
885
- value: 5.0
886
  - type: precision_at_100
887
  value: 0.8750000000000001
888
  - type: precision_at_1000
@@ -1201,7 +1201,7 @@ model-index:
1201
  - type: map_at_5
1202
  value: 16.448999999999998
1203
  - type: mrr_at_1
1204
- value: 71.0
1205
  - type: mrr_at_10
1206
  value: 77.68599999999999
1207
  - type: mrr_at_100
@@ -1225,7 +1225,7 @@ model-index:
1225
  - type: ndcg_at_5
1226
  value: 46.317
1227
  - type: precision_at_1
1228
- value: 71.0
1229
  - type: precision_at_10
1230
  value: 34.4
1231
  - type: precision_at_100
@@ -1370,7 +1370,7 @@ model-index:
1370
  - type: ndcg_at_100
1371
  value: 43.832
1372
  - type: ndcg_at_1000
1373
- value: 47.0
1374
  - type: ndcg_at_3
1375
  value: 33.694
1376
  - type: ndcg_at_5
@@ -2337,19 +2337,19 @@ model-index:
2337
  - type: map_at_5
2338
  value: 1.185
2339
  - type: mrr_at_1
2340
- value: 94.0
2341
  - type: mrr_at_10
2342
- value: 97.0
2343
  - type: mrr_at_100
2344
- value: 97.0
2345
  - type: mrr_at_1000
2346
- value: 97.0
2347
  - type: mrr_at_3
2348
- value: 97.0
2349
  - type: mrr_at_5
2350
- value: 97.0
2351
  - type: ndcg_at_1
2352
- value: 89.0
2353
  - type: ndcg_at_10
2354
  value: 82.30499999999999
2355
  - type: ndcg_at_100
@@ -2361,17 +2361,17 @@ model-index:
2361
  - type: ndcg_at_5
2362
  value: 86.05199999999999
2363
  - type: precision_at_1
2364
- value: 94.0
2365
  - type: precision_at_10
2366
- value: 87.0
2367
  - type: precision_at_100
2368
  value: 63.38
2369
  - type: precision_at_1000
2370
  value: 23.498
2371
  - type: precision_at_3
2372
- value: 94.0
2373
  - type: precision_at_5
2374
- value: 92.0
2375
  - type: recall_at_1
2376
  value: 0.242
2377
  - type: recall_at_10
@@ -2602,20 +2602,32 @@ model-index:
2602
  value: 86.65166462876266
2603
  - type: max_f1
2604
  value: 79.07408783532733
 
2605
  ---
2606
 
2607
- # nomic-embed-text-v1: A Reproducible Long Context (8192) Text Embedder
2608
 
2609
- `nomic-embed-text-v1` is 8192 context length text encoder that surpasses OpenAI text-embedding-ada-002 and text-embedding-3-small performance on short and long context tasks.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2610
 
2611
 
2612
 
2613
- | Name | SeqLen | MTEB | LoCo | Jina Long Context | Open Weights | Open Training Code | Open Data |
2614
- | :-------------------------------:| :----- | :-------- | :------: | :---------------: | :-----------: | :----------------: | :---------- |
2615
- | nomic-embed-text-v1 | 8192 | **62.39** |**85.53** | 54.16 | ✅ | ✅ | ✅ |
2616
- | jina-embeddings-v2-base-en | 8192 | 60.39 | 85.45 | 51.90 | ✅ | ❌ | ❌ |
2617
- | text-embedding-3-small | 8191 | 62.26 | 82.40 | **58.20** | ❌ | ❌ | ❌ |
2618
- | text-embedding-ada-002 | 8191 | 60.99 | 52.7 | 55.25 | ❌ | ❌ | ❌ |
2619
 
2620
 
2621
  ## Hosted Inference API
 
2
  library_name: sentence-transformers
3
  pipeline_tag: sentence-similarity
4
  tags:
5
+ - feature-extraction
6
+ - sentence-similarity
7
+ - mteb
8
+ - transformers
9
+ - transformers.js
10
  model-index:
11
  - name: epoch_0_model
12
  results:
 
882
  - type: precision_at_1
883
  value: 20.096
884
  - type: precision_at_10
885
+ value: 5
886
  - type: precision_at_100
887
  value: 0.8750000000000001
888
  - type: precision_at_1000
 
1201
  - type: map_at_5
1202
  value: 16.448999999999998
1203
  - type: mrr_at_1
1204
+ value: 71
1205
  - type: mrr_at_10
1206
  value: 77.68599999999999
1207
  - type: mrr_at_100
 
1225
  - type: ndcg_at_5
1226
  value: 46.317
1227
  - type: precision_at_1
1228
+ value: 71
1229
  - type: precision_at_10
1230
  value: 34.4
1231
  - type: precision_at_100
 
1370
  - type: ndcg_at_100
1371
  value: 43.832
1372
  - type: ndcg_at_1000
1373
+ value: 47
1374
  - type: ndcg_at_3
1375
  value: 33.694
1376
  - type: ndcg_at_5
 
2337
  - type: map_at_5
2338
  value: 1.185
2339
  - type: mrr_at_1
2340
+ value: 94
2341
  - type: mrr_at_10
2342
+ value: 97
2343
  - type: mrr_at_100
2344
+ value: 97
2345
  - type: mrr_at_1000
2346
+ value: 97
2347
  - type: mrr_at_3
2348
+ value: 97
2349
  - type: mrr_at_5
2350
+ value: 97
2351
  - type: ndcg_at_1
2352
+ value: 89
2353
  - type: ndcg_at_10
2354
  value: 82.30499999999999
2355
  - type: ndcg_at_100
 
2361
  - type: ndcg_at_5
2362
  value: 86.05199999999999
2363
  - type: precision_at_1
2364
+ value: 94
2365
  - type: precision_at_10
2366
+ value: 87
2367
  - type: precision_at_100
2368
  value: 63.38
2369
  - type: precision_at_1000
2370
  value: 23.498
2371
  - type: precision_at_3
2372
+ value: 94
2373
  - type: precision_at_5
2374
+ value: 92
2375
  - type: recall_at_1
2376
  value: 0.242
2377
  - type: recall_at_10
 
2602
  value: 86.65166462876266
2603
  - type: max_f1
2604
  value: 79.07408783532733
2605
+ license: apache-2.0
2606
  ---
2607
 
2608
+ # nomic-embed-text-v1.5: Resizable Production Embeddings with Matryoshka Representation Learning
2609
 
2610
+ `nomic-embed-text-v1.5` is an improvement upon [Nomic Embed](https://huggingface.co/nomic-ai/nomic-embed-text-v1) that utilizes [Matryoshka Representation Learning](https://arxiv.org/abs/2205.13147) which gives developers the flexibility to trade off the embedding size for a negligible reduction in performance.
2611
+
2612
+
2613
+
2614
+ | Name | SeqLen | Dimension | MTEB |
2615
+ | :-------------------------------:| :----- | :-------- | :------: |
2616
+ | nomic-embed-text-v1 | 8192 | 768 | **62.39** |
2617
+ | nomic-embed-text-v1.5 | 8192 | 768 | 62.28 |
2618
+ | nomic-embed-text-v1.5 | 8192 | 512 | 61.96 |
2619
+ | nomic-embed-text-v1.5 | 8192 | 256 | 61.04 |
2620
+ | nomic-embed-text-v1.5 | 8192 | 128 | 59.34 |
2621
+ | nomic-embed-text-v1.5 | 8192 | 64 | 56.10 |
2622
+ | text-embedding-3-small | 8192 | 1536 | 62.26 |
2623
+ | text-embedding-3-small | 8192 | 512 | 61.60 |
2624
+
2625
+
2626
+
2627
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/607997c83a565c15675055b3/CRnaHV-c2wMUMZKw72q85.png)
2628
 
2629
 
2630
 
 
 
 
 
 
 
2631
 
2632
 
2633
  ## Hosted Inference API