INDEX

Explanations

verb to be

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ynes

-0.30

again

-0.26

OMB

-0.26

ĸ

-0.25

ERGE

-0.24

traits

-0.24

.writerow

-0.24

æµ·ç»µ

-0.23

once

-0.23

 caret

-0.23

POSITIVE LOGITS

 advantage

0.29

rac

0.29

anced

0.27

 advant

0.26

 advantages

0.25

abe

0.25

discussion

0.24

 Advantage

0.24

letic

0.24

rac

0.24

Activations Density 0.002%

verb to be

No Comments

No Known Activations