INDEX

Explanations

words indicating initial beliefs or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

-0.63

 ImportError

-0.55

jsxFileName

-0.54

loài

-0.54

 дописавши

-0.54

chluss

-0.53

 للمعارف

-0.53

demie

-0.51

SourceChecksum

-0.51

UrlResolution

-0.50

POSITIVE LOGITS

 initially

1.15

unächst

1.03

 initial

1.03

Initially

0.97

 Initially

0.97

 inicialmente

0.96

 INITIAL

0.91

 inicial

0.88

 اولیه

0.86

Initial

0.86

Activations Density 0.260%

words indicating initial beliefs or actions

No Comments

No Known Activations