INDEX

Explanations

use

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

èµĦäº§è¯Ħä¼°

-0.27

æ´ĹåĩĢ

-0.25

¡°

-0.24

dest

-0.24

_compress

-0.23

 Banc

-0.23

 ÃŃch

-0.23

 unify

-0.23

Vectorizer

-0.23

[loc

-0.23

POSITIVE LOGITS

use

1.04

ä½¿çĶ¨

0.90

 usage

0.88

Use

0.84

Use

0.78

use

0.77

use

0.77

_use

0.74

.use

0.72

usage

0.67

Activations Density 0.004%

use

No Comments

No Known Activations