File size: 4,821 Bytes
cf9609e
 
 
 
 
 
 
 
 
e641fde
cf9609e
 
 
 
 
 
 
 
 
 
e641fde
cf9609e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e641fde
cf9609e
 
e641fde
 
cf9609e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e641fde
cf9609e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e641fde
cf9609e
 
 
 
 
 
 
 
 
 
e641fde
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
---
pretty_name: Deterministic Random Models
license: other
size_categories:
- n<1K
tags:
- llama
- gemma4
- qwen3
- smollm3
- transformers
- safetensors
- gguf
- synthetic
- conformance
- compatibility-testing
---

# Deterministic Random Models

This dataset contains ten small, deterministic language-model fixtures for
model-format, loader, inference, compatibility, and conformance testing. They
are not trained models and must not be used for language-model quality
evaluation.

All weights are synthetic and deterministically generated. No original model
checkpoint weights are included.

## Cases

| Case | Architecture | Parameters | Hugging Face | GGUF | Notable feature |
|---|---|---:|---|---|---|
| `tinyllama-chat` | Llama | 303,744 | F32 | Q4_K_M | GQA, query/KV ratio 8 |
| `smollm2-instruct` | Llama | 46,320 | F32 | Q4_K_M | GQA, query/KV ratio 3 |
| `mobilellama-chat` | Llama | 9,296 | F32 | Q4_K_M | MHA |
| `minicpm5` | Llama | 1,409,664 | F32 | Q4_K_M | explicit head dimension, multiple EOS IDs |
| `deepseek-coder` | Llama | 9,296 | F32 | Q4_K_M | linear RoPE scaling |
| `hermes3-llama31` | Llama | 86,336 | F32 | Q4_K_M | Llama 3 RoPE scaling |
| `livekit-turn-detector` | Llama | 132,336 | F32 | Q4_K_M | explicit head dimension, GQA |
| `gemma4-random-model` | Gemma 4 | 1,519,168 | BF16 | Q4_0 | five-local/one-global attention schedule |
| `qwen3-random-model` | Qwen 3 | 508,800 | BF16 | Q4_0 | wide Q projection and Q/K head norms |
| `smollm3-random-model` | SmolLM3 | 4,917,504 | BF16 | Q4_0 | three-RoPE/one-no-RoPE layer schedule |

The seven Llama cases are derived from real Hugging Face configuration files by
a preservation-first shrinker. Gemma 4, Qwen 3, and SmolLM3 retain
architecture-specific reduced geometries that preserve important
ratios, tensor inventories, and layer schedules observed in locally downloaded
upstream GGUF models. Published case names use `random-model` rather than
`tiny-model` to avoid collision with a separately maintained TinyModel collection.

## Formats and layout

The Llama cases retain the original dataset layout:

```text
<llama-case>/
|-- package/model.safetensors       # canonical F32 weights
|-- gguf/model-Q4_K_M.gguf
|-- tokenizer/
|-- reference/outputs.safetensors
|-- inputs.safetensors
|-- case.json
|-- provenance.json
|-- source-config.json
|-- shrunk-config.json
|-- config-diff.json
`-- validation.json
```

The architecture-specific cases use:

```text
<random-model-case>/
|-- hf-bf16/
|   |-- config.json
|   |-- model.safetensors
|   |-- tokenizer.json
|   `-- tokenizer_config.json
|-- gguf-q4_0/
|   |-- <case>-Q4_0.gguf
|   `-- quantize.log
|-- reference/
|   |-- inputs.json
|   |-- hf-outputs.safetensors
|   `-- gguf-native.json
|-- CONFIG_DECISION.md
`-- metadata.json
```

`manifest.json` is the machine-readable index of all ten model packages and
their SHA-256 hashes and sizes.

## Synthetic weights and tokenizers

Weights use the `tlfloat::LCG64` recurrence with multiplier
`6364136223846793005`, increment `1442695040888963407`, and ten warm-up steps.
Each case records its seed and provenance.

The reduced models use deterministic 128-token auxiliary vocabularies. These
tokenizers cover token IDs `0..127` and preserve each case's special-token
semantics, but they do not reproduce the linguistic behavior of the original
tokenizer. Explicit token IDs are the primary numerical-test interface.

## GGUF generation and validation

GGUF files were generated with upstream `ggml-org/llama.cpp` commit
`40b740ad05c531b9d57aca6698c3ed553a9e784c`.

Every retained GGUF was loaded through that revision and exercised with direct
token IDs for prefill, cached decode, logit extraction, finite-value checks, and
repeated-execution checks. The effective EOG token set was checked against the
model EOS semantics. Per-case metadata records the actual tensor-type histogram,
hashes, commands, and informational comparison with the corresponding
Transformers reference.

Q4_K_M and Q4_0 are lossy formats. Their logits are not required to equal the
F32 or BF16 reference exactly.

## Reproducibility and scope

The Hugging Face weights, configs, and GGUF outputs for Gemma 4, Qwen 3, and SmolLM3 were
independently regenerated and found byte-identical. The Llama cases retain their
source revisions, source-config hashes, shrink decisions, and generation
provenance in each case directory.

This dataset is not a pretrained-model collection, a model-quality benchmark,
or a reproduction of upstream weights or tokenizers. Source-derived configuration
and metadata files may remain subject to terms of their respective upstream
repositories; consult their recorded provenance before redistribution.

See `REPORT.md`, `GGUF_Q4_K_M_REPORT.json`, and
`ARCHITECTURE_RANDOM_MODELS_REPORT.json` for collection-level summaries.