the repo card

README.md

@@ -1,5 +1,5 @@
 ---
-license: cc-by-nc-sa-4.0
+license: mit
 base_model: AbstractPhil/mini-beatrix-3
 library_name: geolip-alephllm
 tags:
@@ -97,6 +97,7 @@ The readout files carry every number; the results files carry every block of eve
 | qwen3/mse_untrained_o0 (the untrained-copy control: the same recipe on a random-initialization copy of the trunk, seed 0; mounts on such a copy only) | final, 2,000 steps | the random trunk reads the two surfaces alike before any arm (.92-.94); its arm reaches .985: the gauge saturates on an untrained trunk, so the trained trunk's closure is the arm's work | .995-.999 |
 | qwen3/mse_solo_o0 (the solo: the same recipe with the arm alone on the bare trunk, no stage arms underneath) | final, 4,000 steps | 0.956 at block 31, 0.953-0.977 at blocks 8-28 (the arm over the eight: 0.892; 0.952-0.971); leads the arm over the eight at 28 of 32 blocks | lowest 0.9880 (the arm over the eight 0.9822); +0.0003 bits per byte on web text |
 | qwen3/mse_gXA_o0_lamA8 (lambda 8 on the plain rows: the whitened-silence fallback, over the gXA eight) | final, 4,000 steps | 0.881 at block 31, 0.949-0.966 at blocks 8-28 (the seed-A arm over the eight at lambda 2: 0.892; mean over the served blocks 0.954 against 0.958) | lowest 0.9899 (block 31; at lambda 2: 0.9822); leads the lambda-2 arm's silence at 30 of 32 blocks; -0.0000 bits per byte on web text |
+| qwen3/mse_solo_o0_bytes (Qwen3's spelling read at EVERY byte (the segmentation-free variant), seed A, alone on the trunk) | final, 4,000 steps | 0.963 at block 12, 0.948-0.988 at blocks 8-28 (before training 0.69 / 0.50 at blocks 12 / 31) | lowest 0.9947 (block 1); +0.0001 bits per byte on web text |
 | bert/mse_solo_o0 (BERT's WordPiece spelling (lower case, '##' continuations), seed A, alone on the trunk) | final, 4,000 steps | 0.742 at block 31, 0.847-0.915 at blocks 8-28 (before training 0.87 / 0.67 at blocks 12 / 31) | lowest 0.9832 (block 31); +0.0006 bits per byte on web text |
 | clip/mse_solo_o0 (CLIP's spelling (the byte map with word-end marks, lower case), seed A, alone on the trunk) | final, 4,000 steps | 0.784 at block 12, 0.676-0.912 at blocks 8-28 (before training 0.43 / 0.29 at blocks 12 / 31) | lowest 0.9819 (block 31); +0.0003 bits per byte on web text |
 | t5/mse_gXA_o0 (T5's sentence-piece spelling, seed A, over the frozen eight) | final, 4,000 steps | 0.857 at block 31, 0.932-0.958 at blocks 8-28 (before training 0.64 / 0.32 at blocks 12 / 31) | lowest 0.9756 (block 31); +0.0005 bits per byte on web text |
@@ -104,7 +105,3 @@ The readout files carry every number; the results files carry every block of eve
 
 The bar for the gap was .85 at every served block from 12 on; the bar for silence .995 at every block beside +.012 bits per byte.
 The MSE form is the one carried forward.
-
-## Licence
-
-The weights are **non-commercial**: [CC BY-NC-SA 4.0](https://creativecommons.org/licenses/by-nc-sa/4.0/), because the training text includes CHILDES child-directed speech (CC BY-NC-SA, via TalkBank) and Wikipedia / WikiText (CC BY-SA). These add-on modules are trained on the Beatrix core and carry its terms; the code that trains and mounts them is MIT ([AbstractEyes/alephllm](https://github.com/AbstractEyes/alephllm)). The full terms and the list of training sources: [LICENSE.md](LICENSE.md).