INDEX

Explanations

cracking

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_2.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.2.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

esti

-0.26

 parach

-0.25

 conduct

-0.24

men

-0.24

men

-0.24

 crafts

-0.24

estre

-0.24

ature

-0.24

 tran

-0.23

è¡Įéķ¿

-0.23

POSITIVE LOGITS

annel

0.29

äº²

0.27

akis

0.26

Tom

0.25

Tom

0.25

ategory

0.25

 Insp

0.24

osit

0.24

ng

0.24

(Pos

0.24

Activations Density 49.763%

cracking

No Comments

No Known Activations