INDEX

Explanations

jh

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ublic

-0.27

 typical

-0.26

 pros

-0.26

Nos

-0.26

æķĻåŃ¦

-0.25

typ

-0.25

Ê°

-0.24

 Pros

-0.24

 Athe

-0.24

alysis

-0.24

POSITIVE LOGITS

-org

0.26

.LocalDateTime

0.25

vern

0.25

rema

0.24

=T

0.24

UIT

0.23

 Reds

0.23

 @"↵

0.23

RTL

0.23

kaar

0.23

Activations Density 0.895%

jh

No Comments

No Known Activations