INDEX

Explanations

structured data or tabulated information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

InitVars

-0.72

 خارجية

-0.60

TagMode

-0.59

ConstraintMaker

-0.59

WireFormatLite

-0.59

transQ

-0.58

 BorderSide

-0.56

OGND

-0.55

 برانيه

-0.53

weightedMode

-0.53

POSITIVE LOGITS

rungsseite

0.38

 المعيارى

0.36

 supérieures

0.34

ódó

0.32

สติ

0.32

 Langen

0.32

 AssemblyTitle

0.31

 supérieurs

0.31

GIA

0.31

 CallOverrides

0.31

Activations Density 0.064%

structured data or tabulated information

No Comments

No Known Activations