INDEX

Explanations

fire

np_max-act · gemini-2.0-flash

The neuron detects occurrences of the substring “fire” (e.g. in words like fire, fireplace, firefighters, firearm).

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 schl

-0.07

Percent

-0.07

Zac

-0.07

 Uzbek

-0.07

Ug

-0.06

aux

-0.06

 zoom

-0.06

各

-0.06

 Marcel

-0.06

_pag

-0.06

POSITIVE LOGITS

 Fire

0.17

 fire

0.16

Fire

0.14

 fireworks

0.12

 FIRE

0.12

火

0.11

fire

0.11

 fires

0.11

.fire

0.10

.Fire

0.10

Activations Density 0.018%

fire

The neuron detects occurrences of the substring “fire” (e.g. in words like fire, fireplace, firefighters, firearm).

No Comments

No Known Activations

fire

The neuron detects occurrences of the substring “fire” (e.g. in words like fire, fireplace, firefighters, firearm).

No Comments

No Known Activations