Commit History
8 days ago
-
v0.2.6: bb address-autofill (weak-multilingual + regex-hint, 4-layer minilm-pruned)
Trained on -weakmulti3hint (weak multilingual data) from minilm-pruned-notest-relabel, encoder_layers=4, bb context, hint + select_option + input_attributes features. Updated tokenizer (18.5k pruned); removed stale vocab.txt/special_tokens_map.json. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>6820ebe
-
v0.2.5: sbsd5 checkpoint, quantized with reduce_range=True (Bug 2064781)
Verified signatureWeights from Metaflow run sbsd5 (autofill-tiny-supportedv2-bb-tinybert-13ep-lr1e4). NOTE: main previously held the jt4qd checkpoint, so this is a checkpoint change as well as a quantization fix. id2label is identical (66 labels), as is the tokenizer and every architecture parameter. Quantized with --per_channel --reduce_range; see the embed repo commit for the VPMADDUBSW int16 saturation mechanism. Weight range: 0/4682058 over +-64 (max |w| = 64). Previous: 308683 over, max 127. Accuracy on testing-supported.txt (4914 rows), sbsd5 checkpoint: variant kappa acc bal_acc w_f1 fp32 0.876357 0.887057 0.722377 0.885980 q8 per-channel 0.873763 0.884615 0.704006 0.883869 q8 per-channel+reduce 0.874187 0.885022 0.723430 0.884059 reduce_range is better on every metric; balanced accuracy recovers the full gap to fp32 (0.704 -> 0.723), i.e. clipping was mostly hurting rare field types. Mirrors rolf-mozilla/tinybert…cd00055
-
Upload 8 files
Verified signature096fa0e -
Upload 5 files
Verified signature3bb0886
-
initial commit
Unverified signature056ef4b -
initial commit
Verified signature8462881
