Add sgpt-nli-bloom-1b3

.gitattributes

@@ -25,3 +25,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
+

1_Pooling/config.json

@@ -0,0 +1,9 @@
+{
+  "word_embedding_dimension": 2048,
+  "pooling_mode_cls_token": false,
+  "pooling_mode_mean_tokens": false,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false,
+  "pooling_mode_weightedmean_tokens": true,
+  "pooling_mode_lasttoken": false
+}

README.md

@@ -0,0 +1,86 @@
+---
+pipeline_tag: sentence-similarity
+tags:
+- sentence-transformers
+- feature-extraction
+- sentence-similarity
+---
+
+# {MODEL_NAME}
+
+This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 2048 dimensional dense vector space and can be used for tasks like clustering or semantic search.
+
+<!--- Describe your model here -->
+
+## Usage (Sentence-Transformers)
+
+Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
+
+```
+pip install -U sentence-transformers
+```
+
+Then you can use the model like this:
+
+```python
+from sentence_transformers import SentenceTransformer
+sentences = ["This is an example sentence", "Each sentence is converted"]
+
+model = SentenceTransformer('{MODEL_NAME}')
+embeddings = model.encode(sentences)
+print(embeddings)
+```
+
+
+
+## Evaluation Results
+
+<!--- Describe how your model was evaluated -->
+
+For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME})
+
+
+## Training
+The model was trained with the parameters:
+
+**DataLoader**:
+
+`sentence_transformers.datasets.NoDuplicatesDataLoader.NoDuplicatesDataLoader` of length 4403 with parameters:
+```
+{'batch_size': 128}
+```
+
+**Loss**:
+
+`sentence_transformers.losses.MultipleNegativesRankingLoss.MNRLGradCache` 
+
+Parameters of the fit()-Method:
+```
+{
+    "epochs": 1,
+    "evaluation_steps": 440,
+    "evaluator": "sentence_transformers.evaluation.EmbeddingSimilarityEvaluator.EmbeddingSimilarityEvaluator",
+    "max_grad_norm": 1,
+    "optimizer_class": "<class 'transformers.optimization.AdamW'>",
+    "optimizer_params": {
+        "lr": 0.00032
+    },
+    "scheduler": "WarmupLinear",
+    "steps_per_epoch": null,
+    "warmup_steps": 441,
+    "weight_decay": 0.01
+}
+```
+
+
+## Full Model Architecture
+```
+SentenceTransformer(
+  (0): Transformer({'max_seq_length': 75, 'do_lower_case': False}) with Transformer model: BloomModel 
+  (1): Pooling({'word_embedding_dimension': 2048, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': True, 'pooling_mode_lasttoken': False})
+)
+```
+
+## Citing & Authors
+
+<!--- Describe where people can find more information -->

config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "/home/opc/sgpt/biencoder/nli_msmarco/sentence-transformers/bloom-1b3",
+  "apply_residual_connection_post_layernorm": false,
+  "architectures": [
+    "BloomModel"
+  ],
+  "attention_dropout": 0.0,
+  "attention_softmax_in_fp32": true,
+  "bias_dropout_fusion": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "hidden_dropout": 0.0,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "masked_softmax_fusion": true,
+  "model_type": "bloom",
+  "n_embed": 2048,
+  "n_inner": null,
+  "n_layer": 24,
+  "num_attention_heads": 16,
+  "offset_alibi": 100,
+  "pretraining_tp": 2,
+  "seq_length": 4096,
+  "skip_bias_add": true,
+  "skip_bias_add_qkv": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.20.0.dev0",
+  "use_cache": false,
+  "vocab_size": 250880
+}

config_sentence_transformers.json

@@ -0,0 +1,7 @@
+{
+  "__version__": {
+    "sentence_transformers": "2.1.0",
+    "transformers": "4.20.0.dev0",
+    "pytorch": "1.10.2"
+  }
+}

eval/similarity_evaluation_sts-dev_results.csv

@@ -0,0 +1,12 @@
+epoch,steps,cosine_pearson,cosine_spearman,euclidean_pearson,euclidean_spearman,manhattan_pearson,manhattan_spearman,dot_pearson,dot_spearman
+0,440,0.8242313306787753,0.8226813115133954,0.8276069619650719,0.8268759567569194,0.8334862792361277,0.8320821542999498,0.7353693665298873,0.7348598843588485
+0,880,0.8459843086553267,0.8503815475776895,0.842878294673236,0.8451332415984594,0.8480524942302925,0.8501835422367507,0.7474383889678284,0.7496324023372983
+0,1320,0.8552344036003329,0.8603697378784526,0.8360007236193618,0.8411717619962727,0.8416522715227013,0.8464700628057116,0.7299126130844387,0.739072574098811
+0,1760,0.8580401062147619,0.8640839343918373,0.8352734951935745,0.841666317497713,0.8404680775310315,0.8465381160962572,0.7257277248318785,0.7377711461544183
+0,2200,0.8608343119089805,0.866170556828279,0.8348571349832736,0.841525970425769,0.839211238760418,0.8458434307006681,0.727537285544208,0.7416178694496526
+0,2640,0.8609322291807961,0.8669794933295952,0.834098925095663,0.8412185342705436,0.8378885926527198,0.8450370545608037,0.7246544989448657,0.7391210367657075
+0,3080,0.8612335308151773,0.8668760555357917,0.8326572542978878,0.8395897555664348,0.8363905686559461,0.8433466318245757,0.7195257907085693,0.737346286430613
+0,3520,0.8626062277494557,0.8681258749036436,0.8346575503908169,0.841873317640566,0.8380398063817516,0.8453286972493597,0.7197310734973534,0.7381475503287697
+0,3960,0.8624141332696987,0.8678907964495984,0.8341648718112045,0.8419911326684907,0.8374000202256104,0.8451632700889332,0.7212668518554104,0.7392373446751543
+0,4400,0.8618680667686823,0.867442027456624,0.8333086384128641,0.8410478063947154,0.8365619082265696,0.8441108013910831,0.7185411402492852,0.737630633601933
+0,-1,0.8618705483964846,0.867452539919395,0.8333168833290586,0.8410583081813221,0.8365715498213021,0.8441200754188356,0.7185606597944942,0.73761900725939

modules.json

@@ -0,0 +1,14 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  }
+]

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e880c3f13ae9a2912d4077266ac000c9611eebcc1123814d4af63a02a7edb72
+size 6889731343

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 75,
+  "do_lower_case": false
+}

sgpt-nli-bloom-1b3_weightedmean_layer-1_results_average_precision.json

@@ -0,0 +1,7 @@
+{
+    "askubuntu": 57.44,
+    "cqadupstack": 14.18,
+    "twitterpara": 73.99,
+    "scidocs": 74.74,
+    "avg": 55.087500000000006
+}

sgpt-nli-bloom-1b3_weightedmean_layer-1_results_detailed.json

@@ -0,0 +1,66 @@
+{
+    "askubuntu": {
+        "map_askubuntu_title": 57.44,
+        "p@1_askubuntu_title": 56.45,
+        "p@5_askubuntu_title": 42.8,
+        "mrr_askubuntu_title": 70.07
+    },
+    "cqadupstack": {
+        "map@100_cqadupstack_unix": 12.58,
+        "ndcg@10_cqadupstack_unix": 14.17,
+        "map@100_cqadupstack_gaming": 27.68,
+        "ndcg@10_cqadupstack_gaming": 30.06,
+        "map@100_cqadupstack_wordpress": 7.22,
+        "ndcg@10_cqadupstack_wordpress": 8.13,
+        "map@100_cqadupstack_stats": 14.86,
+        "ndcg@10_cqadupstack_stats": 15.8,
+        "map@100_cqadupstack_tex": 8.28,
+        "ndcg@10_cqadupstack_tex": 9.28,
+        "map@100_cqadupstack_english": 16.19,
+        "ndcg@10_cqadupstack_english": 17.94,
+        "map@100_cqadupstack_programmers": 12.47,
+        "ndcg@10_cqadupstack_programmers": 13.4,
+        "map@100_cqadupstack_mathematica": 12.05,
+        "ndcg@10_cqadupstack_mathematica": 13.43,
+        "map@100_cqadupstack_physics": 16.5,
+        "ndcg@10_cqadupstack_physics": 18.28,
+        "map@100_cqadupstack_gis": 18.1,
+        "ndcg@10_cqadupstack_gis": 19.14,
+        "map@100_cqadupstack_webmasters": 9.1,
+        "ndcg@10_cqadupstack_webmasters": 10.34,
+        "map@100_cqadupstack_android": 15.16,
+        "ndcg@10_cqadupstack_android": 17.27,
+        "map@100_cqadupstack_avg": 14.18,
+        "ndcg@10_cqadupstack_avg": 15.6
+    },
+    "twitterpara": {
+        "ap_twitter_twitterurl": 77.05,
+        "spearman_twitter_twitterurl": 71.33,
+        "ap_twitter_pit": 70.92,
+        "spearman_twitter_pit": 53.43,
+        "ap_twitter_avg": 73.99,
+        "spearman_twitter_avg": 62.38
+    },
+    "scidocs": {
+        "map_scidocs_cite_euclidean": 72.36,
+        "ndcg_scidocs_cite_euclidean": 86.55,
+        "map_scidocs_cite_cosine": 72.36,
+        "ndcg_scidocs_cite_cosine": 86.55,
+        "map_scidocs_cocite_euclidean": 75.37,
+        "ndcg_scidocs_cocite_euclidean": 88.13,
+        "map_scidocs_cocite_cosine": 75.37,
+        "ndcg_scidocs_cocite_cosine": 88.13,
+        "map_scidocs_coview_euclidean": 76.56,
+        "ndcg_scidocs_coview_euclidean": 87.9,
+        "map_scidocs_coview_cosine": 76.56,
+        "ndcg_scidocs_coview_cosine": 87.9,
+        "map_scidocs_coread_euclidean": 74.67,
+        "ndcg_scidocs_coread_euclidean": 87.14,
+        "map_scidocs_coread_cosine": 74.67,
+        "ndcg_scidocs_coread_cosine": 87.14,
+        "map_scidocs_euclidean_avg": 74.74,
+        "ndcg_scidocs_euclidean_avg": 87.43,
+        "map_scidocs_cosine_avg": 74.74,
+        "ndcg_scidocs_cosine_avg": 87.43
+    }
+}

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "</s>"}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ab8d6865d6043895799adce910945c14ad5f54737d618dbefbe85e84d1a3436
+size 14500694

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<unk>", "eos_token": "</s>", "bos_token": "<s>", "pad_token": "<pad>", "name_or_path": "/home/opc/sgpt/biencoder/nli_msmarco/sentence-transformers/bloom-1b3", "special_tokens_map_file": null, "tokenizer_class": "PreTrainedTokenizerFast"}