Domain-tuned embeddings with held-out recall@5

1_Pooling/config.json

@@ -0,0 +1,5 @@
+{
+    "embedding_dimension": 384,
+    "pooling_mode": "mean",
+    "include_prompt": true
+}

README.md

@@ -0,0 +1,85 @@
+---
+license: apache-2.0
+language:
+  - en
+tags:
+  - zephyr
+  - sentence-transformers
+  - feature-extraction
+  - sentence-similarity
+  - retrieval
+  - rag
+base_model: sentence-transformers/all-MiniLM-L6-v2
+pipeline_tag: sentence-similarity
+library_name: sentence-transformers
+---
+
+# Zephyr RTOS documentation embeddings
+
+`all-MiniLM-L6-v2` fine-tuned on the Zephyr RTOS documentation, for retrieval over
+that corpus.
+
+- **Comparison Space:** https://huggingface.co/spaces/eoinedge/zephyrproject-embeddings-compare
+- **Index and pipeline:** https://huggingface.co/eoinedge/zephyrproject
+- **Source:** https://github.com/eoinjordan/zephyrproject-rag
+
+## What it is for
+
+The base model is trained on general web text, where "binding" means a contract and "west" is a direction. This copy is
+adapted to Zephyr RTOS's vocabulary so those senses separate.
+
+Use it as a drop-in replacement for the base model when embedding Zephyr RTOS
+documentation:
+
+```python
+from sentence_transformers import SentenceTransformer
+
+model = SentenceTransformer("eoinedge/zephyrproject-docs-embeddings")
+vectors = model.encode(passages, normalize_embeddings=True)
+```
+
+## Training
+
+| | |
+|---|---|
+| Base | `sentence-transformers/all-MiniLM-L6-v2` |
+| Pairs | 6,079 mined from the docs' own heading/body structure |
+| Loss | `MultipleNegativesRankingLoss` (in-batch negatives) |
+| Epochs | 1 |
+| Batch size | 32 |
+| Held-out pairs | 400 |
+
+No hand-labelled data and no synthetic questions from a generator — a section
+heading is a natural query for the body beneath it, and it is already written by
+the documentation authors.
+
+## Results, and how to read them
+
+| Metric | Base | Tuned |
+|---|---|---|
+| recall@5 on held-out pairs | 0.63 | **0.86** |
+
+**That number flatters the model, and you should treat it with suspicion.** The
+evaluation measures the same heading→body relationship the model was trained on,
+so it partly measures whether training converged rather than whether retrieval
+improved.
+
+Spot-checking real questions gives a more mixed picture: some clearly improve,
+several are unchanged, and at least one regressed. Real questions are not phrased
+like section headings, which is exactly the gap this evaluation does not cover.
+
+The comparison Space runs both models on the same query so you can judge for
+yourself rather than trusting the headline.
+
+## Limitations
+
+- One epoch, one seed, no hyperparameter search.
+- Evaluation is aligned with the training objective, as above.
+- Trained on Zephyr RTOS documentation only — it will not help on other corpora and
+  may be worse than the base model there.
+- This adapts the **retriever**, not a generator. It does not teach a coding
+  model anything about Zephyr RTOS; it changes which passages that model is handed.
+
+## Licence
+
+Apache-2.0, matching the base model.

config.json

@@ -0,0 +1,30 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "BertModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": null,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 384,
+  "initializer_range": 0.02,
+  "intermediate_size": 1536,
+  "is_decoder": false,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.12.1",
+  "type_vocab_size": 2,
+  "use_cache": false,
+  "vocab_size": 30522
+}

config_sentence_transformers.json

@@ -0,0 +1,14 @@
+{
+  "__version__": {
+    "pytorch": "2.8.0+cpu",
+    "sentence_transformers": "5.6.0",
+    "transformers": "5.12.1"
+  },
+  "default_prompt_name": null,
+  "model_type": "SentenceTransformer",
+  "prompts": {
+    "document": "",
+    "query": ""
+  },
+  "similarity_fn_name": "cosine"
+}

eval.json

@@ -0,0 +1,9 @@
+{
+  "base_model": "sentence-transformers/all-MiniLM-L6-v2",
+  "train_pairs": 6079,
+  "eval_pairs": 400,
+  "epochs": 1,
+  "batch_size": 32,
+  "recall_at_5_baseline": 0.63,
+  "recall_at_5_tuned": 0.86
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:311b52e7d2d55cbb6f2858d0a9feaa761fa92f9e3adc921cc997817e7653f2af
+size 90864192

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.base.modules.transformer.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.sentence_transformer.modules.pooling.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.sentence_transformer.modules.normalize.Normalize"
+  }
+]

sentence_bert_config.json

@@ -0,0 +1,10 @@
+{
+    "transformer_task": "feature-extraction",
+    "modality_config": {
+        "text": {
+            "method": "forward",
+            "method_output_name": "last_hidden_state"
+        }
+    },
+    "module_output_name": "token_embeddings"
+}

tokenizer_config.json

@@ -0,0 +1,24 @@
+{
+  "backend": "tokenizers",
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "is_local": false,
+  "local_files_only": false,
+  "mask_token": "[MASK]",
+  "max_length": 128,
+  "model_max_length": 256,
+  "never_split": null,
+  "pad_to_multiple_of": null,
+  "pad_token": "[PAD]",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "[SEP]",
+  "stride": 0,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "[UNK]"
+}