Spaces:
Running
Running
Update index.html
Browse files- index.html +1 -12
index.html
CHANGED
@@ -45,7 +45,7 @@
|
|
45 |
<li><strong>K2</strong>: Fine-tuned on broader/general-purpose data</li>
|
46 |
<li><strong>K3</strong>: Fine-tuned for clarity and simplicity</li>
|
47 |
</ul>
|
48 |
-
<p>K stands for <em>Knowledge</em>, with higher numbers representing more advanced training stages
|
49 |
</div>
|
50 |
|
51 |
<div class="section">
|
@@ -70,17 +70,6 @@
|
|
70 |
</ul>
|
71 |
</div>
|
72 |
|
73 |
-
<div class="section">
|
74 |
-
<h2>Training Objectives</h2>
|
75 |
-
<p>All models aim to reach a target training loss that signifies strong generalization ability. Training is monitored using:</p>
|
76 |
-
<ul>
|
77 |
-
<li>Loss convergence</li>
|
78 |
-
<li>Gradient norms</li>
|
79 |
-
<li>Learning rate schedules</li>
|
80 |
-
<li>Evaluation tasks (math, logic, generation)</li>
|
81 |
-
</ul>
|
82 |
-
</div>
|
83 |
-
|
84 |
<div class="section">
|
85 |
<h2>Contact & More</h2>
|
86 |
<p>Developed and maintained by <strong>PingVortex</strong>.</p>
|
|
|
45 |
<li><strong>K2</strong>: Fine-tuned on broader/general-purpose data</li>
|
46 |
<li><strong>K3</strong>: Fine-tuned for clarity and simplicity</li>
|
47 |
</ul>
|
48 |
+
<p>K stands for <em>Knowledge</em>, with higher numbers representing more advanced training stages. <strong>Higher doesn't mean the model has more parameters!</strong></p>
|
49 |
</div>
|
50 |
|
51 |
<div class="section">
|
|
|
70 |
</ul>
|
71 |
</div>
|
72 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
73 |
<div class="section">
|
74 |
<h2>Contact & More</h2>
|
75 |
<p>Developed and maintained by <strong>PingVortex</strong>.</p>
|