INDEX

Explanations

This neuron did not activate on any tokens in the provided samples, so its feature cannot be determined from this data.

oai_token-act-pair · gpt-5 Triggered by @liuliujiujiu05

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.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

 poste

-0.31

å²Ľ

-0.27

mis

-0.27

å¸ĸ

-0.27

vk

-0.25

vk

-0.25

oger

-0.25

 post

-0.25

mk

-0.25

@\

-0.25

POSITIVE LOGITS

èµ¶

0.31

 Hastings

0.28

EIF

0.28

angu

0.26

UC

0.25

ABS

0.25

è¶ķ

0.24

utting

0.24

uts

0.24

 Gaines

0.23

Activations Density 43.106%

No Known Activations

This feature has no known activations.