INDEX

Explanations

scientific terminology and phrases related to methodology and processes

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

story, children, or applications

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

Read

-0.31

対

-0.31

 chargeable

-0.30

помним

-0.30

กำ

-0.30

vs

-0.30

 Concentr

-0.28

पया

-0.28

Mok

-0.28

 feveral

-0.28

POSITIVE LOGITS

StoryboardSegue

0.68

 HasFactory

0.62

vician

0.57

 BoxFit

0.56

setVerticalGroup

0.53

enterOuterAlt

0.52

 protoimpl

0.52

titleMargin

0.51

Хьажоргаш

0.50

 pageContext

0.50

Activations Density 0.574%

scientific terminology and phrases related to methodology and processes

story, children, or applications

No Comments

No Known Activations