INDEX

Explanations

phrases related to ongoing actions or processes

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

follow-up

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Portale

-0.52

 companyId

-0.48

ukone

-0.47

PyExc

-0.47

CodeAttribute

-0.46

Portail

-0.45

 Hadid

-0.44

dientemente

-0.44

getMinutes

-0.44

 currentPage

-0.43

POSITIVE LOGITS

astify

0.65

arangay

0.58

 Wikiseite

0.57

 rigue

0.54

träd

0.54

postsleuth

0.53

 Voulez

0.52

ógio

0.51

ésult

0.51

áky

0.50

Activations Density 0.044%

phrases related to ongoing actions or processes

follow-up

No Comments

No Known Activations