LLMScan — Scan Report

DEMOThis report is public and visible to anyone with the link for 30 days, then deleted. Save privately — create free account

vocab-size-mismatch.gguf

Format: GGUF · Size: 5.2 KB · Scanned: 2026-08-10 · 2 ms

SHA256: 9e7eff45b328d1d03a42993cdec05ec80af0da2d…

Security grade

Score: 86/100

1 MEDIUM 2 LOW

Artifact Security

B / 86

Repository Trust

N/A

local upload

Deployment Confidence

Medium

Artifact Security grades the file. Repository Trust is unavailable for local uploads — no external provenance to verify. Deployment Confidence reflects artifact grade only.

▶ Score breakdown

Base score	100
LOW No license field in GGUF metadata (general.license)	-3
LOW No embedded provenance metadata	-3
MEDIUM Embedding tensor 'token_embd.weight' inferred vocab=32 != KV vocab_size=999	-8
Penalty subtotal	86/100
Final	B / 86

GGUF v3 model | architecture: llama | quantization: F32 | 9 tensors | 3 warnings

Runtime compatibility

Runtime	Status	Quant	Architecture	Context
llama.cpp	✓	✓	✓	—
Ollama	✓	✓	✓	—
LM Studio	✓	✓	✓	—
vLLM	✗	N/A		GGUF not supported — requires safetensors format
HuggingFace Transformers	✗	N/A		GGUF not directly supported — use safetensors or PyTorch checkpoint

Model info

Architecture Neural network family — determines which runtime can load this model	llama
Name	llmscan-labs-toy-vocab-mismatch
Quantization Weight storage format derived from actual tensor dtypes; lower bits = smaller file and faster inference at the cost of accuracy	F32
Parameters ~ Estimated total number of weight values; determines VRAM needed at inference	904
Context length Maximum tokens the model can process in a single prompt+response; affects KV-cache memory	128 tokens
Embedding dim Hidden state dimension (d_model); larger = more expressive but more compute per token	8
FFN dim Feed-forward network intermediate size; typically 2.7–4× embedding_length	16
Layers Number of transformer blocks (depth); more layers = more reasoning capacity	1
Attention heads (Q) Number of query heads; GQA/MQA models use fewer KV heads than Q heads	2
KV heads Key/Value heads per layer; fewer than Q heads = GQA; list = per-layer (hybrid architecture)	2
Vocab size Number of unique tokens the tokenizer knows; affects embedding table size	999
Tokenizer Tokenizer algorithm family (e.g. gpt2 = BPE, llama = SentencePiece)	llama
Chat template Jinja2 prompt template embedded in model; controls how messages are formatted for inference	✗ No
Tensors Total number of weight tensors stored in the file	9
File size Size on disk in megabytes	0.01 MB

Security findings (3)

MEDIUM Statistical Anomaly

Embedding tensor 'token_embd.weight' inferred vocab=32 != KV vocab_size=999

Mismatch between embedding dimensions and KV metadata vocab_size may indicate a tampered, corrupted, or mismatched tokenizer/model pair

→ Verify against the original model repository

LOW Supply Chain

No license field in GGUF metadata (general.license)

Absence of license information makes it impossible to determine usage rights

→ Add general.license to model metadata before distributing; use an SPDX identifier

LOW Supply Chain

No embedded provenance metadata

None of general.author, general.source.url, general.base_model.0, or general.source.huggingface.repository are embedded in the GGUF file. Without any origin signal the model cannot be traced to a trusted source or compared against a reference checkpoint for tampering detection. For files uploaded directly, provenance is especially critical because the scanner cannot infer repository origin from the file alone.

→ Add at least one of: general.author, general.source.huggingface.repository, or general.source.url

Integrity & Structure

SHA-256	9e7eff45b328d1d03a42993cdec05ec80af0da2d3e8adb6bbd0b145c0a964505
SHA-512	a595c0c3bb78031c586390a09365fc272ba905fd9067ac60cdb1dd5f75ba81f51f55e5d319ddfcd6…
File size (bytes)	5,344
Header + KV + tensor directory	1,728 bytes (0.00 MB)
Tensor data region	3,616 bytes (0.00 MB)
Trailing bytes	0
Unaccounted bytes	0 ✓
KV metadata keys	14
Tensors parsed	9
Offsets monotone	✓ OK
Max tensor offset	3,104

DType Mix

DType	Count	Total bytes	Total MB	Avg bits/elem
F32	9	3,616	0.0	32.0

Tensor Preview (first 9)

#	Name	Dims	DType	Offset	Bytes	bits/elem
0	token_embd.weight	[8, 32]	F32	0	1,024	32.0
1	output_norm.weight	[8]	F32	1,024	32	32.0
2	blk.0.attn_q.weight	[8, 8]	F32	1,056	256	32.0
3	blk.0.attn_k.weight	[8, 8]	F32	1,312	256	32.0
4	blk.0.attn_v.weight	[8, 8]	F32	1,568	256	32.0
5	blk.0.attn_output.weight	[8, 8]	F32	1,824	256	32.0
6	blk.0.ffn_gate.weight	[8, 16]	F32	2,080	512	32.0
7	blk.0.ffn_down.weight	[16, 8]	F32	2,592	512	32.0
8	blk.0.ffn_up.weight	[8, 16]	F32	3,104	512	32.0

Scan coverage

Checks performed

✓Magic bytes & GGUF version validation
✓Header parse (tensor count, KV count)
✓KV metadata parse — all key/value pairs
✓Tensor directory parse (names, dims, dtype, offsets)
✓Tensor offset monotonicity
✓Tensor overlap detection
✓Tensor out-of-bounds detection
✓Data section alignment verification
✓Trailing bytes after last tensor
✓Unaccounted bytes (header vs file size)
✓Dtype histogram & quantization mix analysis
✓Embedding tensor size cross-validation
✓Chat template static scan (Jinja2/SSTI patterns)
✓Runtime compatibility heuristics
—External model card / repository context: not available for local upload
—Upstream repository: not checked — not available for direct file upload

Not covered

✗Runtime execution or dynamic analysis
✗Behavioral backdoor detection
✗Weight-level semantic backdoor detection
✗Full upstream repository verification
✗License validation beyond embedded metadata
✗Malware scanning of surrounding repo files
✗Adversarial robustness or alignment audit
✗Training data provenance

ⓘ Static scanning can detect structural anomalies and known malicious patterns. It cannot guarantee absence of risk. Treat results as a security signal, not a formal audit.

Raw scan data (.llmscan JSON)

{
  "compatibility": [
    {
      "arch": true,
      "container": true,
      "context": null,
      "notes": "GGUF v3 container: supported | quant F32: supported",
      "quant": true,
      "runtime": "llama.cpp",
      "supported": true
    },
    {
      "arch": true,
      "container": true,
      "context": null,
      "notes": "GGUF via llama.cpp backend: container and quantization supported",
      "quant": true,
      "runtime": "Ollama",
      "supported": true
    },
    {
      "arch": true,
      "container": true,
      "context": null,
      "notes": "GGUF container and quantization supported; GPU/CPU inference",
      "quant": true,
      "runtime": "LM Studio",
      "supported": true
    },
    {
      "arch": null,
      "container": false,
      "context": null,
      "not_supported_msg": "GGUF not supported \u2014 requires safetensors format",
      "notes": "GGUF not supported; requires safetensors format",
      "quant": false,
      "runtime": "vLLM",
      "supported": false
    },
    {
      "arch": null,
      "container": false,
      "context": null,
      "not_supported_msg": "GGUF not directly supported \u2014 use safetensors or PyTorch checkpoint",
      "notes": "GGUF not supported; requires safetensors or PyTorch checkpoint",
      "quant": false,
      "runtime": "HuggingFace Transformers",
      "supported": false
    }
  ],
  "format": "LLM-SCAN",
  "format_version": "1.0",
  "generator": {
    "name": "llmscan-engine",
    "version": "2.0.0"
  },
  "model": {
    "alignment": 32,
    "arch": "llama",
    "architecture": "llama",
    "basename": null,
    "block_count": 1,
    "bos_token_id": null,
    "chat_template": null,
    "context_length": 128,
    "dtype_histogram": {
      "F32": 9
    },
    "dtype_percentage": {
      "F32": 100.0
    },
    "dtype_stats": {
      "F32": {
        "avg_bits_per_elem": 32.0,
        "bytes": 3616,
        "count": 9,
        "mb": 0.0
      }
    },
    "embedding_length": 8,
    "eos_token_id": null,
    "feed_forward_length": 16,
    "file_size_bytes": 5344,
    "file_size_mb": 0.01,
    "finetune": null,
    "format": "gguf",
    "gguf_version": 3,
    "has_chat_template": false,
    "head_count": 2,
    "head_count_kv": 2,
    "header_kv_dir_bytes": 1728,
    "kv_count": 14,
    "kv_meta": {
      "keys": [
        "general.architecture",
        "general.file_type",
        "general.name",
        "llama.attention.head_count",
        "llama.attention.head_count_kv",
        "llama.attention.layer_norm_rms_epsilon",
        "llama.block_count",
        "llama.context_length",
        "llama.embedding_length",
        "llama.feed_forward_length",
        "tokenizer.ggml.model",
        "tokenizer.ggml.token_type",
        "tokenizer.ggml.tokens",
        "tokenizer.ggml.vocab_size"
      ],
      "kv_count": 14,
      "values_redacted": true
    },
    "license": null,
    "max_tensor_offset": 3104,
    "misaligned_tensors": 0,
    "model_name": "llmscan-labs-toy-vocab-mismatch",
    "offsets_monotone": true,
    "oob_tensor_count": 0,
    "overlapping_tensor_pairs": 0,
    "param_estimate": 904,
    "params_estimate": 904,
    "quantization": "F32",
    "scanner_model_info": {
      "alignment": 32,
      "architecture": "llama",
      "basename": null,
      "block_count": 1,
      "bos_token_id": null,
      "chat_template": null,
      "context_length": 128,
      "data_section_start": 1728,
      "dtype_histogram": {
        "F32": 9
      },
      "dtype_percentage": {
        "F32": 100.0
      },
      "dtype_stats": {
        "F32": {
          "avg_bits_per_elem": 32.0,
          "bytes": 3616,
          "count": 9,
          "mb": 0.0
        }
      },
      "embedding_length": 8,
      "eos_token_id": null,
      "feed_forward_length": 16,
      "file_size_bytes": 5344,
      "file_size_mb": 0.01,
      "finetune": null,
      "has_chat_template": false,
      "head_count": 2,
      "head_count_kv": 2,
      "header_end_offset": 1697,
      "header_kv_dir_bytes": 1728,
      "kv_count": 14,
      "kv_keys": [
        "general.architecture",
        "general.file_type",
        "general.name",
        "llama.attention.head_count",
        "llama.attention.head_count_kv",
        "llama.attention.layer_norm_rms_epsilon",
        "llama.block_count",
        "llama.context_length",
        "llama.embedding_length",
        "llama.feed_forward_length",
        "tokenizer.ggml.model",
        "tokenizer.ggml.token_type",
        "tokenizer.ggml.tokens",
        "tokenizer.ggml.vocab_size"
      ],
      "license": null,
      "magic": "GGUF",
      "max_tensor_offset": 3104,
      "misaligned_tensors": 0,
      "model_name": "llmscan-labs-toy-vocab-mismatch",
      "offsets_monotone": true,
      "oob_tensor_count": 0,
      "overlapping_tensor_pairs": 0,
      "param_estimate": 904,
      "quantization": "F32",
      "sha256": "9e7eff45b328d1d03a42993cdec05ec80af0da2d3e8adb6bbd0b145c0a964505",
      "sha512": "a595c0c3bb78031c586390a09365fc272ba905fd9067ac60cdb1dd5f75ba81f51f55e5d319ddfcd6fe63e76a8b3dc653d46b99363f5cd4cdbbe45a8aff4dd377",
      "size_label": null,
      "tensor_bytes_sum": 3616,
      "tensor_count": 9,
      "tensor_delta_bytes": 1728,
      "tensor_delta_mb": 0.0,
      "tensor_preview": [
        {
          "bits_per_elem": 32.0,
          "byte_len": 1024,
          "dims": [
            8,
            32
          ],
          "dtype_id": 0,
          "dtype_name": "F32",
          "index": 0,
          "name": "token_embd.weight",
          "offset": 0
        },
        {
          "bits_per_elem": 32.0,
          "byte_len": 32,
          "dims": [
            8
          ],
          "dtype_id": 0,
          "dtype_name": "F32",
          "index": 1,
          "name": "output_norm.weight",
          "offset": 1024
        },
        {
          "bits_per_elem": 32.0,
          "byte_len": 256,
          "dims": [
            8,
            8
          ],
          "dtype_id": 0,
          "dtype_name": "F32",
          "index": 2,
          "name": "blk.0.attn_q.weight",
          "offset": 1056
        },
        {
          "bits_per_elem": 32.0,
          "byte_len": 256,
          "dims": [
            8,
            8
          ],
          "dtype_id": 0,
          "dtype_name": "F32",
          "index": 3,
          "name": "blk.0.attn_k.weight",
          "offset": 1312
        },
        {
          "bits_per_elem": 32.0,
          "byte_len": 256,
          "dims": [
            8,
            8
          ],
          "dtype_id": 0,
          "dtype_name": "F32",
          "index": 4,
          "name": "blk.0.attn_v.weight",
          "offset": 1568
        },
        {
          "bits_per_elem": 32.0,
          "byte_len": 256,
          "dims": [
            8,
            8
          ],
          "dtype_id": 0,
          "dtype_name": "F32",
          "index": 5,
          "name": "blk.0.attn_output.weight",
          "offset": 1824
        },
        {
          "bits_per_elem": 32.0,
          "byte_len": 512,
          "dims": [
            8,
            16
          ],
          "dtype_id": 0,
          "dtype_name": "F32",
          "index": 6,
          "name": "blk.0.ffn_gate.weight",
          "offset": 2080
        },
        {
          "bits_per_elem": 32.0,
          "byte_len": 512,
          "dims": [
            16,
            8
          ],
          "dtype_id": 0,
          "dtype_name": "F32",
          "index": 7,
          "name": "blk.0.ffn_down.weight",
          "offset": 2592
        },
        {
          "bits_per_elem": 32.0,
          "byte_len": 512,
          "dims": [
            8,
            16
          ],
          "dtype_id": 0,
          "dtype_name": "F32",
          "index": 8,
          "name": "blk.0.ffn_up.weight",
          "offset": 3104
        }
      ],
      "tokenizer_model": "llama",
      "top_tensors_by_size": [
        {
          "byte_len": 1024,
          "dims": [
            8,
            32
          ],
          "dtype": "F32",
          "n_elem": 256,
          "name": "token_embd.weight"
        },
        {
          "byte_len": 512,
          "dims": [
            8,
            16
          ],
          "dtype": "F32",
          "n_elem": 128,
          "name": "blk.0.ffn_gate.weight"
        },
        {
          "byte_len": 512,
          "dims": [
            16,
            8
          ],
          "dtype": "F32",
          "n_elem": 128,
          "name": "blk.0.ffn_down.weight"
        },
        {
          "byte_len": 512,
          "dims": [
            8,
            16
          ],
          "dtype": "F32",
          "n_elem": 128,
          "name": "blk.0.ffn_up.weight"
        },
        {
          "byte_len": 256,
          "dims": [
            8,
            8
          ],
          "dtype": "F32",
          "n_elem": 64,
          "name": "blk.0.attn_q.weight"
        },
        {
          "byte_len": 256,
          "dims": [
            8,
            8
          ],
          "dtype": "F32",
          "n_elem": 64,
          "name": "blk.0.attn_k.weight"
        },
        {
          "byte_len": 256,
          "dims": [
            8,
            8
          ],
          "dtype": "F32",
          "n_elem": 64,
          "name": "blk.0.attn_v.weight"
        },
        {
          "byte_len": 256,
          "dims": [
            8,
            8
          ],
          "dtype": "F32",
          "n_elem": 64,
          "name": "blk.0.attn_output.weight"
        },
        {
          "byte_len": 32,
          "dims": [
            8
          ],
          "dtype": "F32",
          "n_elem": 8,
          "name": "output_norm.weight"
        }
      ],
      "trailing_bytes": 0,
      "unaccounted_bytes": 0,
      "version": 3,
      "vocab_size": 999
    },
    "sha256": "9e7eff45b328d1d03a42993cdec05ec80af0da2d3e8adb6bbd0b145c0a964505",
    "sha512": "a595c0c3bb78031c586390a09365fc272ba905fd9067ac60cdb1dd5f75ba81f51f55e5d319ddfcd6fe63e76a8b3dc653d46b99363f5cd4cdbbe45a8aff4dd377",
    "size_label": null,
    "tensor_bytes_sum": 3616,
    "tensor_count": 9,
    "tensor_delta_bytes": 1728,
    "tensor_delta_mb": 0.0,
    "tensor_preview": [
      {
        "bits_per_elem": 32.0,
        "byte_len": 1024,
        "dims": [
          8,
          32
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 0,
        "name": "token_embd.weight",
        "offset": 0
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 32,
        "dims": [
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 1,
        "name": "output_norm.weight",
        "offset": 1024
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 2,
        "name": "blk.0.attn_q.weight",
        "offset": 1056
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 3,
        "name": "blk.0.attn_k.weight",
        "offset": 1312
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 4,
        "name": "blk.0.attn_v.weight",
        "offset": 1568
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 5,
        "name": "blk.0.attn_output.weight",
        "offset": 1824
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 512,
        "dims": [
          8,
          16
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 6,
        "name": "blk.0.ffn_gate.weight",
        "offset": 2080
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 512,
        "dims": [
          16,
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 7,
        "name": "blk.0.ffn_down.weight",
        "offset": 2592
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 512,
        "dims": [
          8,
          16
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 8,
        "name": "blk.0.ffn_up.weight",
        "offset": 3104
      }
    ],
    "tokenizer_model": "llama",
    "top_tensors_by_size": [
      {
        "byte_len": 1024,
        "dims": [
          8,
          32
        ],
        "dtype": "F32",
        "n_elem": 256,
        "name": "token_embd.weight"
      },
      {
        "byte_len": 512,
        "dims": [
          8,
          16
        ],
        "dtype": "F32",
        "n_elem": 128,
        "name": "blk.0.ffn_gate.weight"
      },
      {
        "byte_len": 512,
        "dims": [
          16,
          8
        ],
        "dtype": "F32",
        "n_elem": 128,
        "name": "blk.0.ffn_down.weight"
      },
      {
        "byte_len": 512,
        "dims": [
          8,
          16
        ],
        "dtype": "F32",
        "n_elem": 128,
        "name": "blk.0.ffn_up.weight"
      },
      {
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype": "F32",
        "n_elem": 64,
        "name": "blk.0.attn_q.weight"
      },
      {
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype": "F32",
        "n_elem": 64,
        "name": "blk.0.attn_k.weight"
      },
      {
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype": "F32",
        "n_elem": 64,
        "name": "blk.0.attn_v.weight"
      },
      {
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype": "F32",
        "n_elem": 64,
        "name": "blk.0.attn_output.weight"
      },
      {
        "byte_len": 32,
        "dims": [
          8
        ],
        "dtype": "F32",
        "n_elem": 8,
        "name": "output_norm.weight"
      }
    ],
    "trailing_bytes": 0,
    "unaccounted_bytes": 0,
    "values_redacted": true,
    "vocab_size": 999
  },
  "report": {
    "bullets": [
      "GGUF v3 model | architecture: llama | quantization: F32 | 9 tensors | 3 warnings",
      "Size: 0.01 MB",
      "SHA256: 9e7eff45b328d1d0..."
    ],
    "recommendations": [
      "Verify against the original model repository",
      "Add general.license to model metadata before distributing; use an SPDX identifier",
      "Add at least one of: general.author, general.source.huggingface.repository, or general.source.url"
    ],
    "summary": "GGUF v3 model | architecture: llama | quantization: F32 | 9 tensors | 3 warnings"
  },
  "scan_context": {
    "duration_ms": 2,
    "ended_at": "2026-08-10T20:38:48+00:00Z",
    "limits": {
      "sandbox": true,
      "timeout_ms": 60000
    },
    "mode": "static-deep",
    "started_at": "2026-08-10T20:38:48+00:00Z"
  },
  "security": {
    "findings": [
      {
        "category": "Supply Chain",
        "detail": "Absence of license information makes it impossible to determine usage rights",
        "recommendation": "Add general.license to model metadata before distributing; use an SPDX identifier",
        "severity": "LOW",
        "title": "No license field in GGUF metadata (general.license)"
      },
      {
        "category": "Supply Chain",
        "detail": "None of general.author, general.source.url, general.base_model.0, or general.source.huggingface.repository are embedded in the GGUF file. Without any origin signal the model cannot be traced to a trusted source or compared against a reference checkpoint for tampering detection. For files uploaded directly, provenance is especially critical because the scanner cannot infer repository origin from the file alone.",
        "recommendation": "Add at least one of: general.author, general.source.huggingface.repository, or general.source.url",
        "severity": "LOW",
        "title": "No embedded provenance metadata"
      },
      {
        "category": "Statistical Anomaly",
        "detail": "Mismatch between embedding dimensions and KV metadata vocab_size may indicate a tampered, corrupted, or mismatched tokenizer/model pair",
        "recommendation": "Verify against the original model repository",
        "severity": "MEDIUM",
        "title": "Embedding tensor \u0027token_embd.weight\u0027 inferred vocab=32 != KV vocab_size=999"
      }
    ],
    "grade": "B",
    "score": 86
  },
  "source": {
    "filename": "vocab-size-mismatch.gguf",
    "hashes": {
      "sha256": "9e7eff45b328d1d03a42993cdec05ec80af0da2d3e8adb6bbd0b145c0a964505"
    },
    "input_type": "file",
    "size_bytes": 5344
  },
  "tensors": {
    "count": 9,
    "dtype_histogram": {
      "F32": 9
    },
    "dtype_stats": {
      "F32": {
        "avg_bits_per_elem": 32.0,
        "bytes": 3616,
        "count": 9,
        "mb": 0.0
      }
    },
    "integrity": {
      "header_kv_dir_bytes": 1728,
      "misaligned_tensors": 0,
      "offsets_monotone": true,
      "oob_tensor_count": 0,
      "overlapping_tensor_pairs": 0,
      "tensor_bytes_sum": 3616,
      "tensor_delta_bytes": 1728,
      "tensor_delta_mb": 0.0,
      "trailing_bytes": 0,
      "unaccounted_bytes": 0
    },
    "preview": [
      {
        "bits_per_elem": 32.0,
        "byte_len": 1024,
        "dims": [
          8,
          32
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 0,
        "name": "token_embd.weight",
        "offset": 0
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 32,
        "dims": [
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 1,
        "name": "output_norm.weight",
        "offset": 1024
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 2,
        "name": "blk.0.attn_q.weight",
        "offset": 1056
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 3,
        "name": "blk.0.attn_k.weight",
        "offset": 1312
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 4,
        "name": "blk.0.attn_v.weight",
        "offset": 1568
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 256,
        "dims": [
          8,
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 5,
        "name": "blk.0.attn_output.weight",
        "offset": 1824
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 512,
        "dims": [
          8,
          16
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 6,
        "name": "blk.0.ffn_gate.weight",
        "offset": 2080
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 512,
        "dims": [
          16,
          8
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 7,
        "name": "blk.0.ffn_down.weight",
        "offset": 2592
      },
      {
        "bits_per_elem": 32.0,
        "byte_len": 512,
        "dims": [
          8,
          16
        ],
        "dtype_id": 0,
        "dtype_name": "F32",
        "index": 8,
        "name": "blk.0.ffn_up.weight",
        "offset": 3104
      }
    ]
  },
  "tokenizer": {
    "bos_token_id": null,
    "chat_template_chars": null,
    "eos_token_id": null,
    "has_chat_template": false,
    "type": "llama",
    "vocab_size": 999
  }
}