INDEX

Explanations

online text snippets

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-stock

-0.28

grade

-0.27

èĽĭ

-0.26

ç¨·

-0.26

colon

-0.26

vs

-0.25

æĶ¹éĢł

-0.25

 Stock

-0.25

iban

-0.25

chied

-0.25

POSITIVE LOGITS

 repeated

0.30

 coordinated

0.29

 distort

0.29

åĲįä¹ī

0.28

 continued

0.27

 cruise

0.26

 approach

0.25

 distortion

0.25

_REPEAT

0.25

.trip

0.25

Activations Density 0.008%

online text snippets

No Comments

No Known Activations