ARotting's picture
Publish Interactive FP32, binary, and ternary comparison
2eec02e verified
Raw
History Blame Contribute Delete
7.08 kB
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width,initial-scale=1">
<title>Bitforge Precision Lab</title>
<style>
:root { color-scheme: dark; font-family: Inter, ui-sans-serif, system-ui; }
* { box-sizing: border-box; }
body { margin: 0; min-height: 100vh; background: #060817; color: #edf4ff; }
canvas { position: fixed; inset: 0; width: 100%; height: 100%; opacity: .55; }
main { position: relative; z-index: 1; width: min(1080px, 92vw); margin: auto;
padding: 72px 0 96px; }
.eyebrow { color: #73e6ff; letter-spacing: .18em; text-transform: uppercase;
font-size: .75rem; font-weight: 800; }
h1 { font-size: clamp(3rem, 8vw, 7rem); line-height: .9; margin: 14px 0 24px;
background: linear-gradient(120deg,#fff,#74e7ff 55%,#b48cff);
-webkit-background-clip: text; color: transparent; }
.lead { max-width: 760px; color: #b8c7e6; font-size: 1.2rem; line-height: 1.65; }
.actions { display: flex; flex-wrap: wrap; gap: 12px; margin: 30px 0 48px; }
a { color: inherit; }
.button { padding: 12px 18px; border-radius: 999px; text-decoration: none;
background: #eaf8ff; color: #07101c; font-weight: 800; }
.button.alt { background: #171d38cc; color: #dce8ff; border: 1px solid #415078; }
.grid { display: grid; grid-template-columns: 1.1fr .9fr; gap: 20px; }
.card { border: 1px solid #344269; background: #0c1128dd; border-radius: 24px;
padding: 24px; backdrop-filter: blur(18px); box-shadow: 0 24px 80px #0008; }
h2 { margin-top: 0; }
pre { white-space: pre-wrap; word-break: break-word; color: #a9bddf;
max-height: 520px; overflow: auto; }
ul { max-height: 520px; overflow: auto; padding-left: 1.2rem; color: #a9bddf; }
li { margin: 8px 0; }
input { width: 100%; padding: 12px; border-radius: 12px; border: 1px solid #344269;
background: #070b1a; color: white; margin-bottom: 12px; }
@media (max-width: 780px) { .grid { grid-template-columns: 1fr; } }
</style>
</head>
<body>
<canvas id="field"></canvas>
<main>
<div class="eyebrow">Jacob Garcia · Hugging Face Model Foundry</div>
<h1>Bitforge Precision Lab</h1>
<p class="lead">Interactive FP32, binary, and ternary comparison. This showcase backs up the
trained artifacts, measured evaluation, and complete runnable source.</p>
<div class="actions">
<a class="button" href="https://huggingface.co/spaces/ARotting/bitforge-precision-lab/tree/main">Explore every file</a>
<a class="button alt" href="https://huggingface.co/ARotting">View the full foundry</a>
</div>
<div class="grid">
<section class="card">
<h2>Verified project card</h2>
<pre># BitForge 1-bit
BitForge trains and compares a full-precision digit classifier, a strict one-bit
weight model, and a ternary-weight model. The low-bit students use straight-through
quantization during training and learn from both labels and the full-precision
teacher&#x27;s softened output distribution.
The binary deployment artifact stores each matrix weight as one packed sign bit,
plus one floating-point scale per output channel and floating-point biases. The
project reports both classification accuracy and the measured inference payload.
Activations, scales, and biases remain floating point, so this is specifically a
one-bit **matrix-weight** experiment rather than a claim that every operation or
parameter is one bit.
## Verified results
| Variant | Test accuracy | Accuracy change |
| --- | ---: | ---: |
| FP32 teacher | 95.78% | reference |
| Packed binary matrix weights | 94.22% | -1.56 points |
| Ternary matrix weights | 95.11% | -0.67 points |
Each network has 4,810 parameters, including 4,736 matrix weights. The measured
inference payload fell from 19,240 bytes for FP32 parameters to 1,184 bytes for
packed signs, per-channel scales, and biases, a 16.25 times reduction. The `.npz`
container itself is 3,270 bytes because it also carries names, shapes, and archive
metadata. An independent reload of the packed signs reproduced 94.22% accuracy.
## Reproduce
```powershell
uv run python projects/bitforge-1bit/train.py
```
</pre>
<h2>Evaluation snapshot</h2>
<pre>{
&quot;benchmark&quot;: &quot;BitForge 1-bit&quot;,
&quot;parameters_per_variant&quot;: 4810,
&quot;matrix_weight_count&quot;: 4736,
&quot;test&quot;: {
&quot;fp32&quot;: {
&quot;accuracy&quot;: 0.9577777777777777,
&quot;cross_entropy&quot;: 0.1542476937174797
},
&quot;binary_weight&quot;: {
&quot;accuracy&quot;: 0.9422222222222222,
&quot;cross_entropy&quot;: 0.2073044627904892
},
&quot;ternary_weight&quot;: {
&quot;accuracy&quot;: 0.9511111111111111,
&quot;cross_entropy&quot;: 0.19603287428617477
}
},
&quot;storage&quot;: {
&quot;fp32_parameter_payload_bytes&quot;: 19240,
&quot;packed_payload_bytes&quot;: 1184,
&quot;container_bytes&quot;: 3270,
&quot;measured_payload_compression&quot;: 16.25
},
&quot;precision_boundary&quot;: {
&quot;matrix_weights&quot;: &quot;one packed bit in binary variant&quot;,
&quot;scales&quot;: &quot;float32 per output channel&quot;,
&quot;biases&quot;: &quot;float32&quot;,
&quot;activations&quot;: &quot;float32&quot;
}
}</pre>
</section>
<section class="card">
<h2>Backed-up artifact tree</h2>
<input id="filter" placeholder="Filter files…" autocomplete="off">
<ul id="files"><li><code>README.md</code></li>
<li><code>__pycache__/app.cpython-311.pyc</code></li>
<li><code>__pycache__/model.cpython-311.pyc</code></li>
<li><code>__pycache__/packing.cpython-311.pyc</code></li>
<li><code>__pycache__/train.cpython-311.pyc</code></li>
<li><code>app.py</code></li>
<li><code>artifacts/bitforge-1bit/binary_qat.safetensors</code></li>
<li><code>artifacts/bitforge-1bit/binary_weights.npz</code></li>
<li><code>artifacts/bitforge-1bit/evaluation.json</code></li>
<li><code>artifacts/bitforge-1bit/fp32.safetensors</code></li>
<li><code>artifacts/bitforge-1bit/ternary_qat.safetensors</code></li>
<li><code>data/split_manifest.parquet</code></li>
<li><code>model.py</code></li>
<li><code>packing.py</code></li>
<li><code>requirements.txt</code></li>
<li><code>train.py</code></li></ul>
</section>
</div>
</main>
<script>
const canvas=document.querySelector('#field'),ctx=canvas.getContext('2d');
let dots=[];
function resize(){canvas.width=innerWidth;canvas.height=innerHeight;
dots=Array.from({length:90},()=>({x:Math.random()*innerWidth,
y:Math.random()*innerHeight,vx:(Math.random()-.5)*.35,vy:(Math.random()-.5)*.35}));}
function draw(){ctx.clearRect(0,0,canvas.width,canvas.height);
for(const d of dots){d.x=(d.x+d.vx+innerWidth)%innerWidth;
d.y=(d.y+d.vy+innerHeight)%innerHeight;ctx.fillStyle='#65dcff99';
ctx.beginPath();ctx.arc(d.x,d.y,1.4,0,7);ctx.fill();}requestAnimationFrame(draw);}
addEventListener('resize',resize);resize();draw();
document.querySelector('#filter').addEventListener('input',e=>{
const q=e.target.value.toLowerCase();for(const li of document.querySelectorAll('li'))
li.hidden=!li.textContent.toLowerCase().includes(q);});
</script>
</body>
</html>