INDEX

Explanations

references to legal scope and policies

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

applies only to scope

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

 Cows

-0.40

UrlHelper

-0.39

FBI

-0.36

KommentareTeilen

-0.36

orle

-0.36

ueve

-0.35

Conse

-0.35

 ControllerBase

-0.35

ugd

-0.35

sendLog

-0.35

POSITIVE LOGITS

:✨

0.73

DockStyle

0.60

 coverage

0.59

 Coverage

0.58

makeConstraints

0.53

httphttps

0.53

 ModelExpression

0.52

 Scope

0.50

 scope

0.50

 jurisdiction

0.50

Activations Density 0.061%

references to legal scope and policies

applies only to scope

No Comments

No Known Activations