INDEX

Explanations

ery

np_max-act · gemini-2.0-flash

tokens containing the “imag-” root, across both general and technical contexts.

oai_token-act-pair · gpt-5 Triggered by @liuliujiujiu05

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_0.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.0.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

æľīæľºä¼ļ

-0.29

èµĦ

-0.26

æ¶Īè´¹

-0.26

æľŁéĹ´

-0.24

Play

-0.24

 magn

-0.24

éģĵ

-0.24

ä¸įå¦Ĥ

-0.23

ä»Ģä¹Īæĺ¯

-0.23

getDefault

-0.23

POSITIVE LOGITS

navbarSupportedContent

0.24

åĨħé©¬

0.24

downloads

0.24

èĬĻ

0.24

ubyte

0.23

izen

0.23

(inertia

0.23

 outr

0.23

.viewport

0.23

iami

0.23

Activations Density 4.731%

ery

tokens containing the “imag-” root, across both general and technical contexts.

No Comments

No Known Activations