INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_1m/average_l0_107

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

1,048,576

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ItemBackground

-0.88

endphp

-0.88

 myſelf

-0.85

 purpoſe

-0.84

 themſelves

-0.83

 pleaſure

-0.82

 MainAxisSize

-0.82

 ſta

-0.81

 ſtate

-0.80

Geplaatst

-0.80

POSITIVE LOGITS



0.52

Ex

0.49

tagHelperRunner

0.46

 ddelweddau

0.46

])).

0.44

mann

0.42

ex

0.41

jem

0.41

�

0.40

tre

0.40

Activations Density 0.000%

No Known Activations

This feature has no known activations.