INDEX

Explanations

phrases that express personal opinions or impressions

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

expressing opinion or intent

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 gynhyrchwyd

-0.58

 themſelves

-0.57

 itſelf

-0.57

 himſelf

-0.56

 تضيفلها

-0.50

 propOrder

-0.49

 nahilalakip

-0.49

Hochspringen

-0.48

parsedMessage

-0.47

WireFormat

-0.47

POSITIVE LOGITS

 intention

0.49

ConstraintMaker

0.47

 belief

0.46

Rüyada

0.44

 Belief

0.42

 intención

0.41

 resourceCulture

0.40

(=)

0.40

 intenção

0.39

⌋

0.39

Activations Density 0.109%

phrases that express personal opinions or impressions

expressing opinion or intent

No Comments

No Known Activations