INDEX

Explanations

numerical data and comparative statistics related to research findings

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

numerical values with units

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

 lyss

-0.31

eme

-0.31

 isti

-0.29

gas

-0.27

me

-0.27

 dynamics

-0.27

 attention

-0.26

actionMode

-0.26

 planning

-0.26

 Patient

-0.25

POSITIVE LOGITS

parsedMessage

0.71

principalColumn

0.68

 насељу

0.64

EndContext

0.63

gucig

0.61

$_(

0.60

 gyhoeddwyd

0.60

 saveiro

0.59

 Frankrijk

0.59

 transfieras

0.59

Activations Density 0.219%

numerical data and comparative statistics related to research findings

numerical values with units

No Comments

No Known Activations