INDEX

Explanations

but

np_max-act · gemini-2.0-flash

The neuron fires on Russian evaluative adjectives and phrases used to express opinions or judgments (e.g. “отличная,” “неинтересное,” “плохое качество”).

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

択

-0.07

isters

-0.07

,params

-0.07

，↵

-0.06

heroes

-0.06

正

-0.06

.MaxLength

-0.06

_Length

-0.06

ーナ

-0.06

plets

-0.06

POSITIVE LOGITS

 кош

0.07

 Quart

0.06

Route

0.06

EDA

0.06

[.

0.06

 وا

0.06

 охорони

0.06

()='

0.06

 перевір

0.06

.O

0.06

Activations Density 0.004%

but

The neuron fires on Russian evaluative adjectives and phrases used to express opinions or judgments (e.g. “отличная,” “неинтересное,” “плохое качество”).

No Comments

No Known Activations