Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,38 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- trl-internal-testing/dolly-chatml-sft
|
| 5 |
+
- HuggingFaceTB/smoltalk
|
| 6 |
+
- diwank/expertllama-chatml
|
| 7 |
+
- HuggingFaceTB/smollm-corpus
|
| 8 |
+
language:
|
| 9 |
+
- en
|
| 10 |
+
pipeline_tag: text-generation
|
| 11 |
+
tags:
|
| 12 |
+
- dumb
|
| 13 |
+
- Asperger
|
| 14 |
+
- LLM
|
| 15 |
+
- SLM
|
| 16 |
+
---
|
| 17 |
+
|
| 18 |
+
# Asperger-7M: world's smallest LLM?
|
| 19 |
+
|
| 20 |
+
**safetensors size: 31.5 MB, FP32**
|
| 21 |
+
|
| 22 |
+
## benchmarks(0-shot)
|
| 23 |
+
|
| 24 |
+
- MMLU: 22.94%
|
| 25 |
+
- GSM8K(500 problem): 1.2%
|
| 26 |
+
|
| 27 |
+
## the goal
|
| 28 |
+
I'm creating an Asperger Large Language Model. In other words, I made idiot Intelligence.
|
| 29 |
+
|
| 30 |
+
---
|
| 31 |
+
|
| 32 |
+
## model family
|
| 33 |
+
|
| 34 |
+
**Asperger-7M**: SoTA ASPERGER models
|
| 35 |
+
Asperger-2M: Mininal ASPERGER models(coming soon!)
|
| 36 |
+
Asperger-Nano: Micro ASPERGER models(coming soon!)
|
| 37 |
+
|
| 38 |
+
auuthor: 56m
|