INDEX

Explanations

references to software toolkits or toolboxes

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

toolkit, toolbox, tools, kit

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

is

-0.46

are

-0.44

<h2>

-0.39

-0.37

Is

-0.36

}${

-0.35

Parcelize

-0.34

<tr>

-0.34

 Enviar

-0.33

Bson

-0.33

POSITIVE LOGITS

Toolkit

1.37

 toolkit

1.21

 Toolkit

1.20

toolkit

1.18

 toolbox

0.97

 Toolbox

0.92

Toolbox

0.88

toolbox

0.85

 tools

0.82

 TOOL

0.81

Activations Density 0.001%

references to software toolkits or toolboxes

toolkit, toolbox, tools, kit

No Comments

No Known Activations