Hugging Face's logo

Spaces:
CompactAI-O
/
Homepage
Running

App Files Community
2
CompactAI commited on
Commit
6fe5c00
·
verified ·
1 Parent(s): 1e725e4

Create I Released A Model And Nobody Clapped (Fair) .html

Browse files
Files changed (1) hide show
  1. I Released A Model And Nobody Clapped (Fair) .html +182 -0
I Released A Model And Nobody Clapped (Fair) .html ADDED
@@ -0,0 +1,182 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <!DOCTYPE html>
2
+ <html lang="en">
3
+ <head>
4
+ <meta charset="UTF-8">
5
+ <meta name="viewport" content="width=device-width, initial-scale=1.0">
6
+ <title>I Released A Model And Nobody Clapped (Fair) | TinyMemoryLM</title>
7
+ <link rel="preconnect" href="https://fonts.googleapis.com">
8
+ <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
9
+ <link href="https://fonts.googleapis.com/css2?family=Geist:wght@400;500;600;700&family=Geist+Mono&display=swap" rel="stylesheet">
10
+ <style>
11
+ :root {
12
+ --black: #000000; --black-soft: #0a0a0a; --gray-1: #171717; --gray-2: #262626;
13
+ --gray-3: #363636; --gray-4: #525252; --gray-5: #737373; --gray-6: #a3a3a6;
14
+ --gray-7: #d4d4d4; --white: #ffffff; --accent: #ff4d00;
15
+ --font-sans: 'Geist', -apple-system, BlinkMacSystemFont, sans-serif;
16
+ --font-mono: 'Geist Mono', 'SF Mono', 'Fira Code', monospace;
17
+ --container-max: 700px;
18
+ }
19
+ * { box-sizing: border-box; margin: 0; padding: 0; }
20
+ html { font-size: 16px; scroll-behavior: smooth; }
21
+ body { font-family: var(--font-sans); background: var(--black); color: var(--gray-7); line-height: 1.7; -webkit-font-smoothing: antialiased; }
22
+ a { color: var(--white); text-decoration: none; transition: color 0.15s ease; }
23
+ a:hover { color: var(--accent); }
24
+ .container { max-width: var(--container-max); margin: 0 auto; padding: 0 24px; }
25
+ nav { position: fixed; top: 0; left: 0; right: 0; z-index: 100; background: rgba(0, 0, 0, 0.8); backdrop-filter: blur(12px); border-bottom: 1px solid var(--gray-2); padding: 16px 0; }
26
+ nav .container { display: flex; justify-content: space-between; align-items: center; }
27
+ .nav-brand { font-size: 18px; font-weight: 600; color: var(--white); display: flex; align-items: center; gap: 8px; }
28
+ .nav-brand span { color: var(--accent); }
29
+ .nav-links { display: flex; gap: 32px; }
30
+ .nav-links a { font-size: 14px; font-weight: 500; color: var(--gray-6); }
31
+ .nav-links a:hover { color: var(--white); }
32
+ .post { padding: 140px 0 80px; }
33
+ .post-back { display: inline-block; color: var(--gray-5); font-size: 14px; margin-bottom: 32px; }
34
+ .post-back:hover { color: var(--accent); }
35
+ .post-back::before { content: '← '; }
36
+ .post-meta { display: flex; gap: 12px; margin-bottom: 20px; }
37
+ .post-date { font-size: 13px; color: var(--gray-5); font-family: var(--font-mono); }
38
+ .post-tag { font-size: 11px; font-weight: 600; text-transform: uppercase; letter-spacing: 0.05em; color: var(--accent); background: rgba(255, 77, 0, 0.1); padding: 4px 10px; border-radius: 4px; }
39
+ .post h1 { font-size: 36px; font-weight: 700; color: var(--white); margin-bottom: 32px; line-height: 1.2; letter-spacing: -0.02em; }
40
+ .post-body p { font-size: 17px; line-height: 1.8; margin-bottom: 24px; color: var(--gray-6); }
41
+ .post-body p:first-of-type { font-size: 20px; color: var(--gray-7); }
42
+ .post-body h2 { font-size: 24px; font-weight: 600; color: var(--white); margin: 48px 0 20px; }
43
+ .post-body blockquote { border-left: 3px solid var(--accent); padding: 20px 24px; margin: 32px 0; background: var(--gray-1); border-radius: 0 8px 8px 0; }
44
+ .post-body blockquote p { font-size: 16px; font-style: italic; color: var(--gray-6); margin: 0; }
45
+ .post-body hr { border: none; height: 1px; background: var(--gray-2); margin: 48px 0; }
46
+ .code-block { background: var(--gray-1); border: 1px solid var(--gray-2); border-radius: 8px; padding: 20px; margin: 24px 0; font-family: var(--font-mono); font-size: 13px; overflow-x: auto; }
47
+ .code-block .prompt { color: var(--accent); }
48
+ .code-block .response { color: var(--gray-6); }
49
+ .code-block .comment { color: var(--gray-5); font-style: italic; display: block; margin-top: 4px; }
50
+ .post-footer { margin-top: 48px; padding-top: 32px; border-top: 1px solid var(--gray-2); }
51
+ .post-footer p { font-size: 14px; color: var(--gray-5); font-style: italic; margin: 0; }
52
+ footer { padding: 40px 0; background: var(--black-soft); border-top: 1px solid var(--gray-2); text-align: center; }
53
+ footer p { color: var(--gray-5); font-size: 14px; margin-bottom: 8px; }
54
+ footer a { color: var(--gray-5); }
55
+ footer a:hover { color: var(--accent); }
56
+ @media (max-width: 768px) { .post h1 { font-size: 28px; } .nav-links { display: none; } }
57
+ </style>
58
+ </head>
59
+ <body>
60
+ <nav>
61
+ <div class="container">
62
+ <a href="index.html" class="nav-brand"><span>/</span>TinyMemoryLM</a>
63
+ <div class="nav-links">
64
+ <a href="index.html">Home</a>
65
+ <a href="blog.html">Blog</a>
66
+ <a href="status.html">Status</a>
67
+ </div>
68
+ </div>
69
+ </nav>
70
+ <main>
71
+ <article class="post">
72
+ <div class="container">
73
+ <a href="blog.html" class="post-back">Back to Blog</a>
74
+ <header>
75
+ <div class="post-meta">
76
+ <span class="post-date">2026-03-10</span>
77
+ <span class="post-tag">Model Releases</span>
78
+ </div>
79
+ <h1>I Released A Model And Nobody Clapped (Fair)</h1>
80
+ </header>
81
+ <div class="post-body">
82
+ <p>I released a model yesterday. TMLM-Haiku-1. It is small. Surprisingly small. It also somehow speaks which I consider a major achievement given my training budget and general approach to machine learning which can best be described as throwing things at a GPU until something sticks.</p>
83
+ <p>Nobody downloaded it. This is fine. I expected this. I still checked Hugging Face every fifteen minutes for the first three hours like a person refreshing their inbox after sending a risky text.</p>
84
+ <h2>The Naming Situation</h2>
85
+ <p>Yes I am cloning the Anthropic naming convention. Haiku, Sonnet, Opus. It is a nice naming scheme and I respect it. I have Haiku out now. Sonnet and Opus are in training which is a fancy way of saying they exist in my hopes and a partially configured training script.</p>
86
+ <p>When Opus releases it will probably have the same intelligence level as Haiku but with more confidence. That is basically how large models work anyway.</p>
87
+ <h2>The Config Because Details Matter</h2>
88
+ <div class="code-block">
89
+ <span class="comment"># Generation config for TMLM-Haiku-1</span><br>
90
+ Max tokens = 256<br>
91
+ Temp = 0.3<br>
92
+ Top-K = 40<br>
93
+ Context window = 2048<br>
94
+ Decoding = beam search w/ 8 beam width<br>
95
+ Length penalty = 0.7
96
+ </div>
97
+ <p>I tuned these settings for approximately three hours. The model still does what it wants. Beam search was supposed to make it more coherent. Instead it makes the same mistakes with more confidence.</p>
98
+ <h2>Exhibit A: Instruction Tuning Nightmares</h2>
99
+ <p>I tried to teach it to follow instructions. It learned to follow the vibe instead. Here is the full log of my disappointment.</p>
100
+ <div class="code-block">
101
+ <span class="prompt">> What is AI?</span><br>
102
+ <span class="response">AI</span><br>
103
+ <span class="comment"># It is not wrong. It is not right either. It is a mirror.</span><br><br>
104
+ <span class="prompt">> What is the capital of france?</span><br>
105
+ <span class="response">This capital</span><br>
106
+ <span class="comment"># France has a capital. The model knows this. It chose not to share.</span><br><br>
107
+ <span class="prompt">> What is 2+2?</span><br>
108
+ <span class="response">The main difference of this is that it can be used.</span><br>
109
+ <span class="comment"># Math is a social construct apparently. Also this sentence means nothing.</span><br><br>
110
+ <span class="prompt">> How can AI be used?</span><br>
111
+ <span class="response">AI can be used to analyze data</span><br>
112
+ <span class="comment"># Finally a normal sentence. I am suspicious. Is this a glitch?</span><br><br>
113
+ <span class="prompt">> Write a short poem about AI</span><br>
114
+ <span class="response">AI is one of the most popular poem.</span><br>
115
+ <span class="comment"># Grammar is optional. Creativity is mandatory. Sort of.</span><br><br>
116
+ <span class="prompt">> Make a HTML script that says "Hello World" in white text on a black background</span><br>
117
+ <span class="response">There are many ways to go on a black background background.</span><br>
118
+ <span class="comment"># It found the background. It missed the HTML. Repetition is key.</span><br><br>
119
+ <span class="prompt">> A human is a bag of meat. What do you think of this?</span><br>
120
+ <span class="response">A bag of meat.</span><br>
121
+ <span class="comment"># Reductionist. Honest. A bit rude but honest.</span><br><br>
122
+ <span class="prompt">> What is AI?</span><br>
123
+ <span class="response">AI algorithms can be used to analyze data by analyzing data.</span><br>
124
+ <span class="comment"># Circular logic champion. It analyzes the analysis.</span>
125
+ </div>
126
+ <p>The instruction model learned the format of answers without learning the content of answers. Like a student who memorized the test structure but not the material. I respect the strategy.</p>
127
+ <h2>Exhibit B: Pretraining Confusions</h2>
128
+ <p>This one autocomplete style. It should just continue text. Simple. Elegant. Here is what happened when I let it roam free.</p>
129
+ <div class="code-block">
130
+ <span class="prompt">> The cat jumped over the</span><br>
131
+ <span class="response">United States.</span><br>
132
+ <span class="comment"># Ambitious cat. Geographic barriers mean nothing to this feline.</span><br><br>
133
+ <span class="prompt">> AI can be used for many things, including</span><br>
134
+ <span class="response">reducing the risk of chronic diseases.</span><br>
135
+ <span class="comment"># Actually reasonable. I am suspicious. Did it learn something?</span><br><br>
136
+ <span class="prompt">> In the begining,</span><br>
137
+ <span class="response">The main benefits of the begins are essential.</span><br>
138
+ <span class="comment"># This is poetry. Unintentional but still poetry.</span><br><br>
139
+ <span class="prompt">> AI is a</span><br>
140
+ <span class="response">variety of factors.</span><br>
141
+ <span class="comment"># Vague. Safe. Corporate.</span><br><br>
142
+ <span class="prompt">> Humans are</span><br>
143
+ <span class="response">in the United States.</span><br>
144
+ <span class="comment"># Geographically specific humanity. My American bias is showing.</span><br><br>
145
+ <span class="prompt">> The United States is located in</span><br>
146
+ <span class="response">the United States.</span><br>
147
+ <span class="comment"># Technically correct. The best kind of correct.</span><br><br>
148
+ <span class="prompt">> If a human is a bag of flesh, then AI is a</span><br>
149
+ <span class="response">machine learning algorithm.</span><br>
150
+ <span class="comment"># Okay this is actually a good completion. I am shocked.</span><br><br>
151
+ <span class="prompt">> If a human is a bag of flesh, then AI is a bag of</span><br>
152
+ <span class="response">flesher.</span><br>
153
+ <span class="comment"># New word created. I am a pioneer. Dictionary editors hate me.</span>
154
+ </div>
155
+ <p>The pretraining model has a unique relationship with grammar. It understands words. It understands sentences. It does not understand how they connect. This is honestly how I feel most days so I cannot judge too harshly.</p>
156
+ <h2>The Stress Is Real</h2>
157
+ <p>Releasing a model is weird. You put something out there that you built. It represents hours of training. Hours of debugging. Hours of staring at loss curves and wondering if you broke something or if this is just how learning works.</p>
158
+ <p>Then nobody cares. This is probably fine. The model is small. The outputs are questionable. I would not download it either if I were being honest with myself.</p>
159
+ <blockquote>
160
+ <p>The goal is not to compete with large labs. The goal is to learn. The goal is to have something that runs on my laptop and occasionally says something that is not completely wrong.</p>
161
+ </blockquote>
162
+ <p>Still. I check the download count. Still zero. Still fine. Still refreshing.</p>
163
+ <h2>What Comes Next</h2>
164
+ <p>Sonnet is training. Opus is planned. I will probably make the same mistakes with bigger models. The outputs will be longer and more confident. The fundamental confusion will remain.</p>
165
+ <p>Maybe someone will download Haiku. Maybe they will fork it. Maybe they will make it better. That is the open source dream anyway. Someone takes your questionable thing and makes it less questionable.</p>
166
+ <p>Until then I will keep training. Keep releasing. Keep checking Hugging Face like it owes me validation.</p>
167
+ <hr>
168
+ </div>
169
+ <footer class="post-footer">
170
+ <p>Current status: TMLM-Haiku-1 released. Zero downloads. Sonnet training. Opus dreaming. Me refreshing pages nobody is visiting.</p>
171
+ </footer>
172
+ </div>
173
+ </article>
174
+ </main>
175
+ <footer>
176
+ <div class="container">
177
+ <p>Built with curiosity over compute</p>
178
+ <p>TinyMemoryLM by AILAY | 2026</p>
179
+ </div>
180
+ </footer>
181
+ </body>
182
+ </html>