INDEX

Explanations

segments of text that conclude or introduce new sections or ideas

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

IntoConstraints

-0.69

AnimationsModule

-0.65

Rhestr

-0.58



-0.56

новниш

-0.56

makeConstraints

-0.56

 surla

-0.54

脚注の使い方

-0.52

 Chwiliwch

-0.52

 otomatig

-0.52

POSITIVE LOGITS

SharedCtor

0.49

bootstrapcdn

0.36

XMLSchema

0.36

 Autorisations

0.35

 disambiguazione

0.32

erdings

0.31

+#+

0.31

 Marín

0.30

Institute

0.30

 elegido

0.30

Activations Density 0.072%

segments of text that conclude or introduce new sections or ideas

No Comments

No Known Activations