INDEX

Explanations

where

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_7.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.7.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

Mon

-0.27

Mon

-0.26

 ...↵

-0.25

 monot

-0.25

 exchange

-0.25

å°±æĥ³

-0.24

mon

-0.24

 ...↵↵

-0.24

['$

-0.24

ante

-0.24

POSITIVE LOGITS

çĥ«

0.26

-bin

0.26

æĲĤ

0.26

 Positions

0.25

_positions

0.25

 positions

0.25

stairs

0.25

ä¸Ĭæ¬¡

0.25

elper

0.24

positions

0.24

Activations Density 0.001%

where

No Comments

No Known Activations