INDEX

Explanations

will

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

é£İåĲ¹

-0.29

.named

-0.28

å¾ĢæĿ¥

-0.26

 minded

-0.26

ä¿ĿæĮģçĿĢ

-0.26

 reversal

-0.25

çİĩè¾¾åĪ°

-0.24

 warranted

-0.23

breaking

-0.23

wealth

-0.23

POSITIVE LOGITS

not

0.40

 also

0.33

be

0.32

 hopefully

0.29

 probably

0.28

not

0.28

 still

0.28

 definitely

0.27

NOT

0.27

Activations Density 0.146%

will

No Comments

No Known Activations