INDEX

Explanations

occurrences of statistical results and associated p-values

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

references and figures

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

LookAnd

-0.54

MessageTagHelper

-0.50

AddTagHelper

-0.49

hisattva

-0.47

########.

-0.47

Toponymie

-0.46

 AssemblyCulture

-0.45

 pleaſure

-0.45

gjenge

-0.45

 faſt

-0.44

POSITIVE LOGITS

indd

0.49

 CanadaChoose

0.48

>−</

0.47

"-

0.46

("-

0.46

 صوتيه

0.44

—-

0.44

('-

0.44

="-

0.43

»-

0.42

Activations Density 0.039%

occurrences of statistical results and associated p-values

references and figures

No Comments

No Known Activations