INDEX

Explanations

references to program state and validity checks

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

ьажоргаш

-0.50



-0.50

ViewFeatures

-0.49

bolista

-0.46

 []).

-0.45

+#+#

-0.45

 Chwiliwch

-0.45

LookAnd

-0.44

 HasFactory

-0.42

فحة

-0.41

POSITIVE LOGITS

istoitu

0.41

setMaximum

0.40

 mineures

0.39

ggable

0.38

 factors

0.38

 מלא

0.35

 зас

0.34

 ones

0.34

 successes

0.33

vänt

0.33

Activations Density 0.063%

references to program state and validity checks

No Comments

No Known Activations