INDEX

Explanations

does

np_max-act · gemini-2.0-flash

phrases indicating negations or exceptions, particularly words like "not," "without," "but," and "never" that introduce contrasting or limiting statements.

oai_token-act-pair · deepseek-v3 Triggered by @finnwit

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

prs

-0.28

nx

-0.27

zig

-0.25

 Wait

-0.25

rapid

-0.24

rg

-0.24

ryo

-0.23

IGNORE

-0.23

.gdx

-0.23

Suz

-0.23

POSITIVE LOGITS

æĹ¢

0.30

èĲ½åľ°

0.29

ä¸įé«ĺ

0.29

PELL

0.27

 creat

0.26

ismatic

0.25

ssh

0.25

ä¸Ĭæ¦ľ

0.25

ä¸įè¿ĩ

0.24

å±±ä¸ľ

0.24

Activations Density 0.010%

does

phrases indicating negations or exceptions, particularly words like "not," "without," "but," and "never" that introduce contrasting or limiting statements.

No Comments

No Known Activations

does

phrases indicating negations or exceptions, particularly words like "not," "without," "but," and "never" that introduce contrasting or limiting statements.

No Comments

No Known Activations