{
 "schema_version": "1.7.0",
 "id": "x_NXMC-2024-0027",
 "modified": "2026-10-19T02:33:38.839Z",
 "published": "2026-10-18T09:00:52.565Z",
 "summary": "jina-embeddings-v2-base-code: structural change at a594cdd (2024-02-05)",
 "details": "jina-embeddings-v2-base-code: EOS declared as \"</s>\" (was unset); generation that relied on the library default now stops on this token; BOS declared as \"<s>\" (was unset); vocab_size changed from 30528 to 61056; the embedding and output head change shape, so adapters and merges made against the old vocabulary no longer load cleanly (2024-02-05).\n\nRepository: https://huggingface.co/jinaai/jina-embeddings-v2-base-code. Revisions from a594cdd8ed6d6310f9dcada354356340e025cfb9 onward carry this change; a pin at a08d5882d94bfcc69188a4533e5c3bb1f197d37a or earlier does not.\n\nChanged keys:\n- dtype: float32 -> float16 (structural)\n- vocab_size: 30528 -> 61056 (structural)\n- tokenizer bos_token: unset -> <s> (tokenizer)\n- tokenizer eos_token: unset -> </s> (tokenizer)\n- tokenizer model_max_length: 2147483648 -> 8192 (tokenizer)\n- tokenizer pad_token: [PAD] -> <pad> (tokenizer)\n- tokenizer tokenizer_class: BertTokenizer -> RobertaTokenizer (tokenizer)\n- tokenizer unk_token: [UNK] -> <unk> (tokenizer)\n\nThis record says what the change does mechanically. It is not a report that anything broke for anyone.",
 "affected": [
  {
   "ranges": [
    {
     "type": "GIT",
     "repo": "https://huggingface.co/jinaai/jina-embeddings-v2-base-code",
     "events": [
      {
       "introduced": "a594cdd8ed6d6310f9dcada354356340e025cfb9"
      }
     ]
    }
   ],
   "database_specific": {
    "model": "jinaai/jina-embeddings-v2-base-code",
    "previous_revision": "a08d5882d94bfcc69188a4533e5c3bb1f197d37a"
   }
  }
 ],
 "references": [
  {
   "type": "INTRODUCED",
   "url": "https://huggingface.co/jinaai/jina-embeddings-v2-base-code/commit/a594cdd8ed6d6310f9dcada354356340e025cfb9"
  },
  {
   "type": "PACKAGE",
   "url": "https://huggingface.co/jinaai/jina-embeddings-v2-base-code"
  },
  {
   "type": "WEB",
   "url": "https://www.neurarch.com/changes-feeds/models/jinaai/jina-embeddings-v2-base-code.html"
  }
 ],
 "database_specific": {
  "source": "neurarch model change tracker",
  "model": "jinaai/jina-embeddings-v2-base-code",
  "kind": "structural",
  "classes": [
   "structural",
   "tokenizer"
  ],
  "changed_keys": [
   {
    "file": "config",
    "key": "dtype",
    "before": "float32",
    "after": "float16",
    "class": "structural"
   },
   {
    "file": "config",
    "key": "vocab_size",
    "before": 30528,
    "after": 61056,
    "class": "structural"
   },
   {
    "file": "tokenizer",
    "key": "bos_token",
    "before": null,
    "after": "<s>",
    "class": "tokenizer"
   },
   {
    "file": "tokenizer",
    "key": "eos_token",
    "before": null,
    "after": "</s>",
    "class": "tokenizer"
   },
   {
    "file": "tokenizer",
    "key": "model_max_length",
    "before": 2147483648,
    "after": 8192,
    "class": "tokenizer"
   },
   {
    "file": "tokenizer",
    "key": "pad_token",
    "before": "[PAD]",
    "after": "<pad>",
    "class": "tokenizer"
   },
   {
    "file": "tokenizer",
    "key": "tokenizer_class",
    "before": "BertTokenizer",
    "after": "RobertaTokenizer",
    "class": "tokenizer"
   },
   {
    "file": "tokenizer",
    "key": "unk_token",
    "before": "[UNK]",
    "after": "<unk>",
    "class": "tokenizer"
   }
  ],
  "impact": {
   "memory": true,
   "serving": true,
   "tokenizer": true,
   "licence": null
  },
  "breaking": true,
  "breaking_rule": "https://www.neurarch.com/changes-feeds/advisories/index.json#breaking_rule",
  "consequence": {
   "params": {
    "before": 114772992,
    "after": 138218496
   },
   "weightsGbBf16": {
    "before": 0.23,
    "after": 0.276
   },
   "trainPeakGb": {
    "before": 23.042,
    "after": 23.492
   },
   "fingerprint": null,
   "context": null,
   "eos": {
    "before": null,
    "after": "</s>"
   },
   "bos": {
    "before": null,
    "after": "<s>"
   }
  },
  "commit_date": "2024-02-05T07:13:55.000Z",
  "detected_at": "2026-10-18T09:00:52.565Z",
  "detected_by": "backfill",
  "days_since_first_commit": 79,
  "license": "CC-BY-4.0"
 }
}
