INDEX

Explanations

tale

np_max-act · gemini-2.0-flash

The neuron activates on the phrase “tale of,” i.e. it detects when “tale” is followed by “of.”

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

 ""},↵

-0.07

.checkbox

-0.07

}}],↵

-0.07

ический

-0.07

:");
↵

-0.07

 fontFamily

-0.07

buttonShape

-0.07

 borderWidth

-0.07

ScreenWidth

-0.07

.Dispatch

-0.06

POSITIVE LOGITS

 tale

0.13

 Tale

0.13

 tales

0.12

 Tales

0.10

 babes

0.07

 dile

0.07

tre

0.06

ale

0.06

 Vale

0.06

번

0.06

Activations Density 0.006%

tale

The neuron activates on the phrase “tale of,” i.e. it detects when “tale” is followed by “of.”

No Comments

No Known Activations