INDEX

Explanations

information about product features and enhancements

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

preceding "features"

np_acts-logits-general · gemini-2.0-flash

new features and enhancements

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

SPATH

-0.55

änden

-0.51

gdx

-0.51

 Italijanski

-0.50

 AssemblyCompany

-0.50

 rechange

-0.47

rhosis

-0.46

 Infórmanos

-0.45

IBOutlet

-0.45

illots

-0.44

POSITIVE LOGITS

 features

2.75

 feature

2.39

 Features

2.35

features

2.32

Features

2.17

feature

2.08

 FEATURES

2.06

 Feature

2.02

 functionality

1.97

Feature

1.95

Activations Density 0.522%

information about product features and enhancements

preceding "features"

new features and enhancements

No Comments

No Known Activations