INDEX

Explanations

say "ir"

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_32.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

çī¹æ´¾åĳĺ

-0.27

orgh

-0.25

hor

-0.25

.named

-0.25

obr

-0.24

çĤĢ

-0.23

edom

-0.23

_Str

-0.23

è¡·

-0.23

äº²æīĭ

-0.23

POSITIVE LOGITS

ir

0.35

Ir

0.33

ir

0.32

Ir

0.31

IR

0.30

IRA

0.27

(ir

0.27

IR

0.26

fix

0.26

irmed

0.26

Activations Density 0.000%

say "ir"

No Comments

No Known Activations