google/electra-small-discriminator
Original13.5Mparameters
Fixture13.5Mparameters
Layer stacks1compared
Findings0to review
✓ Same architecture, smaller
Every module class, layer kind and checked property of the original is in the fixture.
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.
electra.encoder.layer
Original · 12
Fixture · 12
| Kind | Original | Fixture | |
|---|---|---|---|
| Layer | 12 | 12 | ✓ 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
| Component | Original | Share | Fixture | Share |
|---|---|---|---|---|
| electra.encoder.layer | 9.5M | 69.9% | 9.5M | 69.9% |
| electra.embeddings | 4.0M | 29.3% | 4.0M | 29.3% |
| discriminator_predictions.dense | 65.8K | 0.5% | 65.8K | 0.5% |
| electra.embeddings_project | 33.0K | 0.2% | 33.0K | 0.2% |
| discriminator_predictions.dense_prediction | 257 | 0.0% | 257 | 0.0% |
Preserved properties
| Config | Property | Original | Fixture | |
|---|---|---|---|---|
| model | model_type | electra | electra | ✓ kept |
| model | vocab_size | 30522 | 30522 | ✓ kept |
| model | tie_word_embeddings | True | True | ✓ kept |
| model | hidden_act | gelu | gelu | ✓ kept |
| model | max_position_embeddings | 512 | 512 | ✓ kept |
Module classes
Instances of each class in the built model (prefix Electra omitted).
| Class | Original | Fixture |
|---|---|---|
| Linear | 75 | 75 |
| Dropout | 37 | 37 |
| LayerNorm | 25 | 25 |
| GELUActivation | 13 | 13 |
| Attention | 12 | 12 |
| Intermediate | 12 | 12 |
| Layer | 12 | 12 |
| Output | 12 | 12 |
| SelfAttention | 12 | 12 |
| SelfOutput | 12 | 12 |
| Embedding | 3 | 3 |
| DiscriminatorPredictions | 1 | 1 |
| Embeddings | 1 | 1 |
| Encoder | 1 | 1 |
| ForPreTraining | 1 | 1 |
| Model | 1 | 1 |
| ModuleList | 1 | 1 |
Config changes
0 fields changed, 35 unchanged.
| Field | Original | Fixture |
|---|