INDEX

Explanations

punctuation

np_max-act · gemini-2.0-flash

Tokens that appear before or after punctuation marks, often closing parentheses, commas, or end of statements.

eleuther_acts_top20 · deepseek-v3 Triggered by @finnwit

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_9.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.9.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ä¸ĭæīĭ

-0.27

 Grace

-0.25

 Matching

-0.25

edException

-0.24

.matches

-0.24

è¤ļ

-0.24

Grace

-0.24

å§¿

-0.24

æ»ĭ

-0.24

å±±æ°´

-0.24

POSITIVE LOGITS

raries

0.28

ala

0.27

ahir

0.26

obar

0.25

alien

0.25

Ã©ny

0.25

 comet

0.25

alin

0.24

armor

0.24

ola

0.24

Activations Density 7.143%

punctuation

Tokens that appear before or after punctuation marks, often closing parentheses, commas, or end of statements.

No Comments

No Known Activations