Update README.md
Browse files
README.md
CHANGED
|
@@ -29,7 +29,7 @@ This repo contains the Base (completed pretrain cudalm checkpoint) variant of LS
|
|
| 29 |
LS3-Micro-Base has a 24.9% higher Intelligence Index than LS-Micro Base, while having 13% less total parameters. It also nearly matches LS2.5-108M-A17M with almost 11x less parameters.
|
| 30 |
|
| 31 |
- Intelligence Index was calculated using the formula from [Open SLM Leaderboard](https://huggingface.co/spaces/AxiomicLabs/Open_SLM_Leaderboard). (*Results reproducable via Cudalm's built in bench tool, see bottom quickstart*)
|
| 32 |
-
-
|
| 33 |
|
| 34 |
---
|
| 35 |
|
|
|
|
| 29 |
LS3-Micro-Base has a 24.9% higher Intelligence Index than LS-Micro Base, while having 13% less total parameters. It also nearly matches LS2.5-108M-A17M with almost 11x less parameters.
|
| 30 |
|
| 31 |
- Intelligence Index was calculated using the formula from [Open SLM Leaderboard](https://huggingface.co/spaces/AxiomicLabs/Open_SLM_Leaderboard). (*Results reproducable via Cudalm's built in bench tool, see bottom quickstart*)
|
| 32 |
+
- Intelligence density derived using the following formula: Intelligence Index / sqrt(Active Params (millions) * Total Params (millions))
|
| 33 |
|
| 34 |
---
|
| 35 |
|