INDEX

Explanations

punctuation and formatting marks in text

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

يكب

-0.48

TintMode

-0.46

StoryboardSegue

-0.45

 indígen

-0.42

 Affili

-0.41

IndentedString

-0.41

erialization

-0.40

ſy

-0.40

AccessorTable

-0.39

 northeast

-0.39

POSITIVE LOGITS

etc

0.68

etc

0.53

Etc

0.50

 sebagainya

0.47

Etc

0.44

issory

0.43

などは

0.43

usw

0.42

等等

0.41

opress

0.41

Activations Density 0.106%

punctuation and formatting marks in text

No Comments

No Known Activations