{
  "id": "xgen-7b-8k-base",
  "name": "xgen-7b-8k-base",
  "identifiers": {
    "huggingface": [
      "Salesforce/xgen-7b-8k-base"
    ]
  },
  "developer": "Salesforce AI Research",
  "release_date": {
    "value": "2023-09-07",
    "source": "https://huggingface.co/Salesforce/xgen-7b-8k-base",
    "status": "recorded"
  },
  "weights_status": "open",
  "availability": {
    "value": "available",
    "checked": "2026-09-24",
    "source": "https://huggingface.co/Salesforce/xgen-7b-8k-base"
  },
  "license": {
    "value": "apache-2.0",
    "source": "https://huggingface.co/Salesforce/xgen-7b-8k-base/blob/14ca1dadca23f5fb99d99c90b1638ecff2a52f9e/README.md",
    "status": "partial",
    "note": "From HF card metadata (uploader-declared); confirm against the license text."
  },
  "architecture": {
    "family": "decoder_only",
    "note": "family read from config 'architectures': ['LlamaForCausalLM'].",
    "n_layers": {
      "value": 32,
      "source": "https://huggingface.co/Salesforce/xgen-7b-8k-base/raw/14ca1dadca23f5fb99d99c90b1638ecff2a52f9e/config.json",
      "status": "recorded"
    },
    "hidden_size": {
      "value": 4096,
      "source": "https://huggingface.co/Salesforce/xgen-7b-8k-base/raw/14ca1dadca23f5fb99d99c90b1638ecff2a52f9e/config.json",
      "status": "recorded"
    },
    "n_heads": {
      "value": 32,
      "source": "https://huggingface.co/Salesforce/xgen-7b-8k-base/raw/14ca1dadca23f5fb99d99c90b1638ecff2a52f9e/config.json",
      "status": "recorded"
    },
    "vocab_size": {
      "value": 51200,
      "source": "https://huggingface.co/Salesforce/xgen-7b-8k-base/raw/14ca1dadca23f5fb99d99c90b1638ecff2a52f9e/config.json",
      "status": "recorded"
    },
    "context_length": {
      "value": 8192,
      "source": "https://huggingface.co/Salesforce/xgen-7b-8k-base/raw/14ca1dadca23f5fb99d99c90b1638ecff2a52f9e/config.json",
      "status": "recorded"
    },
    "positional_encoding": {
      "value": null,
      "status": "not_recorded"
    }
  },
  "training_data": {
    "value": null,
    "status": "not_recorded",
    "note": "Salesforce's paper/card do not disclose the pretraining corpus composition."
  },
  "techniques": [],
  "primary_sources": [
    "https://huggingface.co/Salesforce/xgen-7b-8k-base",
    "https://huggingface.co/Salesforce/xgen-7b-8k-base/raw/14ca1dadca23f5fb99d99c90b1638ecff2a52f9e/config.json"
  ],
  "record_history": [
    {
      "date": "2026-09-24",
      "change": "ingested as candidate from HF (Salesforce/xgen-7b-8k-base@14ca1dadca23f5fb99d99c90b1638ecff2a52f9e)",
      "by": "ingest_hf.py"
    },
    {
      "date": "2026-09-24",
      "change": "preparer: developer, release date set",
      "by": "claude (preparer, Sonnet 5)"
    },
    {
      "date": "2026-09-24",
      "change": "preparer cleanup: removed 2 stale flag(s)",
      "by": "claude (preparer, Sonnet 5)"
    },
    {
      "date": "2026-09-24",
      "change": "preparer cleanup: training_data flag reworded to reflect a real, checked gap rather than an unfilled TODO",
      "by": "claude (preparer, Sonnet 5)"
    },
    {
      "date": "2026-09-24",
      "change": "reviewed and promoted from staging (0 edge(s) accepted)",
      "by": "Wilson Pruitt"
    }
  ]
}
