INDEX

Explanations

negation expressions, particularly related to logical conditions

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

 canst

-0.44

setcounter

-0.44

Ball

-0.43

Swagger

-0.42

geführt

-0.42

 barbe

-0.41

 Shakspeare

-0.41

mml

-0.40

zuführen

-0.40

()%

-0.38

POSITIVE LOGITS

(!(

1.31

(!(

1.13

!(

0.91

(!_

0.68

(!$

0.68

脚注の使い方

0.64

!(

0.63

 autorytatywna

0.63

 noDo

0.61

 disambiguazione

0.61

Activations Density 0.008%

negation expressions, particularly related to logical conditions

No Comments

No Known Activations