Add Consight v2 embeddings: bge-base fine-tuned with hard negatives

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c0a174815082ca24aa5ccb5a56c24417dce19d6cac2d27da1c9d2036df435a3
+size 437951328

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

real_world_results.json

@@ -0,0 +1,11 @@
+{
+  "test_pairs": 89,
+  "baseline_model": "all-MiniLM-L6-v2",
+  "trained_model": "/home/ubuntu/sweep/data/model_sweep/bge-base_lr5e6_ep5_bs16",
+  "baseline_mean": 0.513227381021454,
+  "trained_mean": 0.6044960109872765,
+  "improvement": 0.09126862996582241,
+  "improvement_pct": 17.783273718595147,
+  "distraction_mean_margin": 0.17822580886621997,
+  "distraction_positive_pct": 75.34246575342466
+}

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+    "max_seq_length": 512,
+    "do_lower_case": true
+}

special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

sweep_result.json

@@ -0,0 +1,38 @@
+{
+  "config_name": "bge-base_lr5e6_ep5_bs16",
+  "model": "BAAI/bge-base-en-v1.5",
+  "hyperparameters": {
+    "learning_rate": 5e-06,
+    "epochs": 5,
+    "batch_size": 16,
+    "warmup_steps": 100,
+    "max_grad_norm": 0.5
+  },
+  "training_time_seconds": 294.1,
+  "synthetic_results": {
+    "baseline_model": "all-MiniLM-L6-v2",
+    "trained_model": "/home/ubuntu/sweep/data/model_sweep/bge-base_lr5e6_ep5_bs16",
+    "sample_size": 200,
+    "baseline_mean": 0.6809912909008563,
+    "trained_mean": 0.7275207257270813,
+    "improvement_mean": 0.04652943482622496,
+    "improvement_pct": 6.832603507259692,
+    "baseline_median": 0.7112560570240021,
+    "trained_median": 0.7321392893791199,
+    "improvement_median": 0.020883232355117798,
+    "trained_std": 0.1384458249346926,
+    "trained_min": 0.406196266412735,
+    "trained_max": 0.9855192303657532
+  },
+  "real_world_results": {
+    "test_pairs": 89,
+    "baseline_model": "all-MiniLM-L6-v2",
+    "trained_model": "/home/ubuntu/sweep/data/model_sweep/bge-base_lr5e6_ep5_bs16",
+    "baseline_mean": 0.513227381021454,
+    "trained_mean": 0.6044960109872765,
+    "improvement": 0.09126862996582241,
+    "improvement_pct": 17.783273718595147,
+    "distraction_mean_margin": 0.17822580886621997,
+    "distraction_positive_pct": 75.34246575342466
+  }
+}

tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

training_config.json

@@ -0,0 +1,16 @@
+{
+  "base_model": "BAAI/bge-base-en-v1.5",
+  "pairs_file": "/home/ubuntu/sweep/data/pairs_combined.jsonl",
+  "distractors_file": "/home/ubuntu/sweep/data/distractors_combined.jsonl",
+  "loss_function": "MultipleNegativesRankingLoss",
+  "hard_negatives": true,
+  "epochs": 5,
+  "batch_size": 16,
+  "warmup_steps": 100,
+  "learning_rate": 5e-06,
+  "max_grad_norm": 0.5,
+  "device": null,
+  "max_pairs": null,
+  "timestamp": "2026-02-18T18:08:27.651228",
+  "completion_time": "2026-02-18T18:13:21.789188"
+}

training_results.json

@@ -0,0 +1,15 @@
+{
+  "baseline_model": "all-MiniLM-L6-v2",
+  "trained_model": "/home/ubuntu/sweep/data/model_sweep/bge-base_lr5e6_ep5_bs16",
+  "sample_size": 200,
+  "baseline_mean": 0.6809912909008563,
+  "trained_mean": 0.7275207257270813,
+  "improvement_mean": 0.04652943482622496,
+  "improvement_pct": 6.832603507259692,
+  "baseline_median": 0.7112560570240021,
+  "trained_median": 0.7321392893791199,
+  "improvement_median": 0.020883232355117798,
+  "trained_std": 0.1384458249346926,
+  "trained_min": 0.406196266412735,
+  "trained_max": 0.9855192303657532
+}