docs: drop internal benchmark path from the GGUF verification note
Browse files
README.md
CHANGED
|
@@ -579,8 +579,8 @@ m.eval()
|
|
| 579 |
`gguf_read.q4_blob(path)` returns the carried `xlit.q4` bytes from the compact file: 588,891 bytes, byte-identical to
|
| 580 |
the original (sha256 `652eafe0c1e2b46fec85663825621284f0a5911f2d7cc6e13b1182ce47cd9803`).
|
| 581 |
|
| 582 |
-
**Verification.** Both files were run back through the reference decoder on a fixed sample of 400 words (the first 400
|
| 583 |
-
|
| 584 |
dictionary pack, comparing against the shipped `xlit.q4`:
|
| 585 |
|
| 586 |
* `aksharalp-1m-v2-q4.gguf`: identical top-1, identical candidate sets and identical score-ranked candidate lists for
|
|
|
|
| 579 |
`gguf_read.q4_blob(path)` returns the carried `xlit.q4` bytes from the compact file: 588,891 bytes, byte-identical to
|
| 580 |
the original (sha256 `652eafe0c1e2b46fec85663825621284f0a5911f2d7cc6e13b1182ce47cd9803`).
|
| 581 |
|
| 582 |
+
**Verification.** Both files were run back through the reference decoder on a fixed sample of 400 words (the first 400 distinct alphabetic romanised words of the
|
| 583 |
+
Aksharantar Hindi test split), with the
|
| 584 |
dictionary pack, comparing against the shipped `xlit.q4`:
|
| 585 |
|
| 586 |
* `aksharalp-1m-v2-q4.gguf`: identical top-1, identical candidate sets and identical score-ranked candidate lists for
|