INDEX

Explanations

1

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.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

é¸£

-0.27

åĳľ

-0.26

son

-0.26

åĵ¨

-0.26

 ways

-0.26

oga

-0.25

son

-0.25

 sons

-0.25

nock

-0.25

oft

-0.24

POSITIVE LOGITS

º«

0.32

ç´§ç´§

0.25

peror

0.25

 Indo

0.25

 asympt

0.24

å®Įç¾İçļĦ

0.24

è¿ĳæĹ¥

0.23

åµ¬

0.23

 remin

0.23

 exposition

0.23

Activations Density 0.081%

1

No Comments

No Known Activations