← MIC coverage of Transformers model types

facebook/s2t-small-librispeech-asr

huggingface-mic/s2t-small-librispeech-asr-mic@88acd7a6239d ← facebook/s2t-small-librispeech-asr@256e91a603c0 · Speech2TextForConditionalGeneration · random weights

CI: no failure on main in 30 days · MIC failures dashboard

Original29.5Mparameters
Fixture1.5Mparameters
Layer stacks2compared
Findings0to review

✓ Same architecture, smaller

Every module class, layer kind and checked property of the original is in the fixture.

Model tester (advisory)

The model's own small test config in its ModelTester (23 literal values, 23 set by the model's own tester, the rest generic base-tester defaults). Testers are usually representative but not always, so these notes are for review, not rules.

The fixture covers every per-layer kind the tester sets, and shrinks every size it does.

Layer stacks

One square per layer, colored by kind: the layer's module classes and type attributes. Hover a square for its index and kind.

model.encoder.layers

Original · 12
Fixture · 2
KindOriginalFixture
EncoderLayer122✓ kept

model.decoder.layers

Original · 6
Fixture · 2
KindOriginalFixture
DecoderLayer62✓ kept

Where the parameters are

Share of each model's parameters per component. Vocabulary and patch sizes are kept, so embeddings usually take a larger share of the fixture.

Original
Fixture
ComponentOriginalShareFixtureShare
model.encoder.layers15.8M53.4%66.9K4.3%
model.decoder.layers9.5M32.1%100.5K6.5%
model.decoder2.6M8.7%640.1K41.4%
model.encoder1.7M5.8%738.6K47.8%

Preserved properties

ConfigPropertyOriginalFixture
modelmodel_typespeech_to_textspeech_to_text✓ kept
modelvocab_size1000010000✓ kept
modeltie_word_embeddingsTrueTrue✓ kept

Module classes

Instances of each class in the built model (prefix Speech2Text omitted).

ClassOriginalFixture
Linear13333
LayerNorm4412
Attention246
ReLU184
EncoderLayer122
DecoderLayer62
ModuleList33
Conv1d22
SinusoidalPositionalEmbedding22
Conv1dSubsampler11
Embedding11
Decoder11
Encoder11
ForConditionalGeneration11
Model11

Config changes

5 fields changed, 35 unchanged.

FieldOriginalFixture
d_model25664
decoder_ffn_dim2048128
decoder_layers62
encoder_ffn_dim2048128
encoder_layers122