INDEX

Explanations

technical/code-related texts

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

å½±

-0.36

 shadow

-0.32

::-

-0.31

 styling

-0.29

Shadow

-0.28

çŃĶåį·

-0.28

èĭĳ

-0.28

é»¥

-0.28

 Shadow

-0.28

shadow

-0.27

POSITIVE LOGITS

ateur

0.31

ateurs

0.28

.cent

0.26

æ¡´

0.26

[Test

0.26

ovo

0.26

(access

0.26

 pedest

0.25

é²µ

0.25

globals

0.25

Activations Density 0.000%

technical/code-related texts

No Comments

No Known Activations