Update README.md
Browse files
README.md
CHANGED
|
@@ -9,6 +9,8 @@ tags:
|
|
| 9 |
|
| 10 |
# **I made Qwen2-0.5B run 1000× smaller with no matrix multiplications and almost no RAM usage — adaptive sparse selection at inference time, and it actually works.**
|
| 11 |
|
|
|
|
|
|
|
| 12 |
**Status: experimental / work in progress.** This is a research test, not a production release, not a compression method, and not a claim of a new architecture. Numbers below are placeholders — real plots and stats will be added once benchmark runs are complete.Currently, according to tests, it doesn't use any RAM at all, but there may be errors. With this compression, we compressed by about 30%, but the main matrices are not multiplied.
|
| 13 |
|
| 14 |
## What this actually is
|
|
|
|
| 9 |
|
| 10 |
# **I made Qwen2-0.5B run 1000× smaller with no matrix multiplications and almost no RAM usage — adaptive sparse selection at inference time, and it actually works.**
|
| 11 |
|
| 12 |
+
# contact: Twitter: https://x.com/liberal17th email: bogunusov@gmail.com
|
| 13 |
+
|
| 14 |
**Status: experimental / work in progress.** This is a research test, not a production release, not a compression method, and not a claim of a new architecture. Numbers below are placeholders — real plots and stats will be added once benchmark runs are complete.Currently, according to tests, it doesn't use any RAM at all, but there may be errors. With this compression, we compressed by about 30%, but the main matrices are not multiplied.
|
| 15 |
|
| 16 |
## What this actually is
|