INDEX

Explanations

instances of questioning and seeking clarification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_4/width_65k/average_l0_66

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.4.hook_mlp_out

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

bindValue

-0.62

 autorytatywna

-0.54

 للاسماء

-0.52

سجيل

-0.52

glTexCoord

-0.51

 AttributeSet

-0.50

enterOuterAlt

-0.50

 jSON

-0.50

PAIGN

-0.50

cepcion

-0.50

POSITIVE LOGITS

 linkovi

0.64

physis

0.54

 تانيه

0.54

AndEndTag

0.53

visiae

0.53

zzleHttp

0.52

 مرئيه

0.52

 IndexError

0.52

nthetic

0.52

 continúas

0.52

Activations Density 0.241%

instances of questioning and seeking clarification

No Comments

No Known Activations