INDEX

Explanations

of

np_max-act · gemini-2.0-flash

The neuron flags abstract‐noun tokens that denote potential outcomes or ramifications (e.g. “implications,” “consequences,” “potential”).

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

œur

-0.07

Rac

-0.07

_MUL

-0.07

函数

-0.07

Subset

-0.07

Population

-0.07

 backpage

-0.06

见

-0.06

strpos

-0.06

 залеж

-0.06

POSITIVE LOGITS

Sdk

0.07

(Of

0.06

OBJ

0.06

ожд

0.06

 पर

0.06

_predict

0.06

uci

0.06

fect

0.06

ifications

0.06

ِب

0.05

Activations Density 0.038%

of

The neuron flags abstract‐noun tokens that denote potential outcomes or ramifications (e.g. “implications,” “consequences,” “potential”).

No Comments

No Known Activations