{
 "schema_version": "1.7.0",
 "id": "x_NXMC-2024-0169",
 "modified": "2026-10-19T02:33:38.839Z",
 "published": "2026-10-18T09:00:52.565Z",
 "summary": "saiga_llama3_8b: tokenizer change at 0af818c (2024-04-19)",
 "details": "saiga_llama3_8b: pad token unset to 128002; batched training masks a different id; generation defaults changed (do_sample unset to true, max_new_tokens unset to 1536, repetition_penalty unset to 1.12, temperature unset to 0.2); a call that does not pass these explicitly now generates differently (2024-04-19).\n\nRepository: https://huggingface.co/IlyaGusev/saiga_llama3_8b. Revisions from 0af818c24e23e1aff7f3ad895c6862f4d3598971 onward carry this change; a pin at f4eb6c87c589f7df4da8618110df56f0d181bad4 or earlier does not.\n\nChanged keys:\n- generation do_sample: unset -> true (generation-defaults)\n- generation max_new_tokens: unset -> 1536 (generation-defaults)\n- generation pad_token_id: unset -> 128002 (tokenizer)\n- generation repetition_penalty: unset -> 1.12 (generation-defaults)\n- generation temperature: unset -> 0.2 (generation-defaults)\n- generation top_k: unset -> 30 (generation-defaults)\n- generation top_p: unset -> 0.9 (generation-defaults)\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/IlyaGusev/saiga_llama3_8b",
     "events": [
      {
       "introduced": "0af818c24e23e1aff7f3ad895c6862f4d3598971"
      }
     ]
    }
   ],
   "database_specific": {
    "model": "IlyaGusev/saiga_llama3_8b",
    "previous_revision": "f4eb6c87c589f7df4da8618110df56f0d181bad4"
   }
  }
 ],
 "references": [
  {
   "type": "INTRODUCED",
   "url": "https://huggingface.co/IlyaGusev/saiga_llama3_8b/commit/0af818c24e23e1aff7f3ad895c6862f4d3598971"
  },
  {
   "type": "PACKAGE",
   "url": "https://huggingface.co/IlyaGusev/saiga_llama3_8b"
  },
  {
   "type": "WEB",
   "url": "https://www.neurarch.com/changes-feeds/models/IlyaGusev/saiga_llama3_8b.html"
  }
 ],
 "database_specific": {
  "source": "neurarch model change tracker",
  "model": "IlyaGusev/saiga_llama3_8b",
  "kind": "tokenizer",
  "classes": [
   "tokenizer",
   "generation-defaults"
  ],
  "changed_keys": [
   {
    "file": "generation",
    "key": "do_sample",
    "before": null,
    "after": true,
    "class": "generation-defaults"
   },
   {
    "file": "generation",
    "key": "max_new_tokens",
    "before": null,
    "after": 1536,
    "class": "generation-defaults"
   },
   {
    "file": "generation",
    "key": "pad_token_id",
    "before": null,
    "after": 128002,
    "class": "tokenizer"
   },
   {
    "file": "generation",
    "key": "repetition_penalty",
    "before": null,
    "after": 1.12,
    "class": "generation-defaults"
   },
   {
    "file": "generation",
    "key": "temperature",
    "before": null,
    "after": 0.2,
    "class": "generation-defaults"
   },
   {
    "file": "generation",
    "key": "top_k",
    "before": null,
    "after": 30,
    "class": "generation-defaults"
   },
   {
    "file": "generation",
    "key": "top_p",
    "before": null,
    "after": 0.9,
    "class": "generation-defaults"
   }
  ],
  "impact": {
   "memory": false,
   "serving": true,
   "tokenizer": true,
   "licence": null
  },
  "breaking": true,
  "breaking_rule": "https://www.neurarch.com/changes-feeds/advisories/index.json#breaking_rule",
  "consequence": {
   "params": null,
   "weightsGbBf16": null,
   "trainPeakGb": null,
   "fingerprint": null,
   "context": null,
   "eos": null,
   "bos": null
  },
  "commit_date": "2024-04-19T11:52:31.000Z",
  "detected_at": "2026-10-18T09:00:52.565Z",
  "detected_by": "backfill",
  "days_since_first_commit": 0,
  "license": "CC-BY-4.0"
 }
}