INDEX

Explanations

references to specific locations, governmental bodies, and cultural contexts

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

 informée

-0.74

 snippetHide

-0.63

 <<<<<<<<<<<<<<

-0.59

 препратки

-0.58

addCriterion

-0.58

ThroughAttribute

-0.57

 estekak

-0.56

RotationOrder

-0.55

 հղումներ

-0.54

SharedDtor

-0.54

POSITIVE LOGITS

 futuros

0.40

 habitudes

0.40

 économies

0.39

 future

0.39

ünf

0.37

 beras

0.37

 précieux

0.37

 matières

0.37

 suivante

0.36

 ejecut

0.36

Activations Density 0.082%

references to specific locations, governmental bodies, and cultural contexts

No Comments

No Known Activations