INDEX

Explanations

quality

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

not

-0.36

Not

-0.31

arat

-0.30

è½»

-0.29

Not

-0.29

_not

-0.29

 clear

-0.28

not

-0.28

.high

-0.27

 High

-0.26

POSITIVE LOGITS

å¾Ĺ

0.46

 best

0.40

best

0.38

-best

0.35

(best

0.35

 BEST

0.35

BEST

0.34

å¾Ĺå¥½

0.33

æľĢä½³

0.33

å¾Ĺå¤ļ

0.33

Activations Density 0.223%

quality

No Comments

No Known Activations