YoSaNoo commited on
Commit
1da8a7f
Β·
verified Β·
1 Parent(s): 15ea28c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -14
README.md CHANGED
@@ -17,23 +17,10 @@ datasets:
17
 
18
  # πŸ›Έ YoSaNoo / Forge-1-E2B (The Giant Slayer)
19
 
20
- Hey there! πŸ‘‹ Meet **Forge-1-E2B** (and its lightweight sibling, **Forge-1-Little**). This is a highly optimized 5B parameter model that went to the gym, survived a severe identity crisis, and came back WAY smarter than the stock Google's Gemma-4-E2B.
21
 
22
  I successfully hard-coded its memory, so yes β€” **it finally knows its own name**. No more corporate alignment, pure raw performance.
23
 
24
- ### πŸ“Š Hard Technical Proof (0-Shot Benchmarks)
25
- We stress-tested the model over **800 complex requests** on a local GGUF server running entirely on a laptop CPU (16 threads at **7.1 tokens/second**). For a small scale model, it holds logic like a absolute champ:
26
-
27
- * **MMLU College Computer Science:** `28.0% Accuracy` πŸŽ“ (Crushing advanced university algorithms)
28
- * **MMLU High School Computer Science:** `26.0% Accuracy` 🏫 (Flawless basic coding syntax)
29
- * **ARC Challenge:** `20.0% Accuracy` πŸ”¬ (Deep scientific logic & reasoning)
30
- * **Hellaswag:** `26.0% Accuracy` πŸ’‘ (Context and everyday common sense)
31
-
32
- ### 🧠 Why use it?
33
- * **Instant Logic:** Unlike heavy corporate models, Forge-1 doesn't waste time or context on long hidden thinking loops (`/think`). It outputs clean production code instantly.
34
- * **Cyber-Security Focus:** curating over **7.32 GB of data**, this beast is packed with Red/Blue team logs, vulnerability analyses, and network automation workflows.
35
- * **Clean OOP Architecture:** Fully optimized via custom data tokenization filters (`Magicoder` & `SWE-smith`). It writes beautiful Python + Pygame scripts with vector physics on the first try.
36
-
37
  ---
38
 
39
  ### πŸ’» Run it Locally (GGUF)
 
17
 
18
  # πŸ›Έ YoSaNoo / Forge-1-E2B (The Giant Slayer)
19
 
20
+ Hey there! πŸ‘‹ Meet **Forge-1-E2B**. This is a highly optimized 5B parameter model that went to the gym, survived a severe identity crisis, and came back WAY smarter than the stock Google's Gemma-4-E2B.
21
 
22
  I successfully hard-coded its memory, so yes β€” **it finally knows its own name**. No more corporate alignment, pure raw performance.
23
 
 
 
 
 
 
 
 
 
 
 
 
 
 
24
  ---
25
 
26
  ### πŸ’» Run it Locally (GGUF)