INDEX

Explanations

rifle

np_max-act · gemini-2.0-flash

references to rifles and closely related firearm terms, especially in military or shooting contexts.

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

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_11.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.11.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

 thaw

-0.27

raise

-0.27

 freeze

-0.27

 bubb

-0.26

&type

-0.26

freeze

-0.26

åĭº

-0.26

 ...↵

-0.26

åŃ£èĬĤ

-0.25

Raised

-0.25

POSITIVE LOGITS

ry

0.29

/raw

0.25

IFS

0.24

åĲĮåŁİ

0.24

wich

0.24

/content

0.24

 organization

0.24

 Credit

0.23

otre

0.23

.observable

0.23

Activations Density 0.369%

rifle

references to rifles and closely related firearm terms, especially in military or shooting contexts.

No Comments

No Known Activations