INDEX

Explanations

references to applications and usage instructions

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

several mentions

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

enterOuterAlt

-0.56

 otomatig

-0.53

PYX

-0.48

 Normdatei

-0.47

 Wiktionnaire

-0.45

 Enix

-0.44

GTCX

-0.44

⟬

-0.43

Demografie

-0.43

newswire

-0.43

POSITIVE LOGITS

伍

0.45

inson

0.44

iteter

0.42

entingan

0.42

ensos

0.42

nstein

0.40

IUrlHelper

0.40

orten

0.40

Behind

0.40

スン

0.39

Activations Density 0.156%

references to applications and usage instructions

several mentions

No Comments

No Known Activations