Update README.md
Browse files
README.md
CHANGED
|
@@ -24,6 +24,8 @@ pipeline_tag: text-generation
|
|
| 24 |
|
| 25 |
# İvme-Conversate-22M-Base
|
| 26 |
|
|
|
|
|
|
|
| 27 |
**İvme** (Turkish: *acceleration*) is a series of stupidly small language models built to punch above their weight. This is the first release: a 22M parameter decoder-only base model trained from scratch on a dense, quality-filtered corpus.
|
| 28 |
|
| 29 |
The goal is not production deployment. The goal is to see how well a sub-25M model can perform when every decision — architecture, data mix, optimizer, training schedule — is made deliberately.
|
|
|
|
| 24 |
|
| 25 |
# İvme-Conversate-22M-Base
|
| 26 |
|
| 27 |
+
|
| 28 |
+
|
| 29 |
**İvme** (Turkish: *acceleration*) is a series of stupidly small language models built to punch above their weight. This is the first release: a 22M parameter decoder-only base model trained from scratch on a dense, quality-filtered corpus.
|
| 30 |
|
| 31 |
The goal is not production deployment. The goal is to see how well a sub-25M model can perform when every decision — architecture, data mix, optimizer, training schedule — is made deliberately.
|