INDEX

Explanations

code

np_max-act · gemini-2.0-flash

The neuron primarily fires on discourse-opening or connective tokens (e.g. “As,” “Sure,” “Here,” “In,” “The”) that signal the start of a response or paragraph.

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

Vict

-0.07

 divisor

-0.07

intendent

-0.07

Kate

-0.07

 melting

-0.06

Danny

-0.06

Max

-0.06

距

-0.06

 weitere

-0.06

POSITIVE LOGITS

appe

0.06

_TUN

0.06

 sapi

0.06

cmb

0.06

tö

0.06

がお

0.06

 Exist

0.06

arse

0.06

[]):

0.06

 Πανεπ

0.06

Activations Density 0.026%

code

The neuron primarily fires on discourse-opening or connective tokens (e.g. “As,” “Sure,” “Here,” “In,” “The”) that signal the start of a response or paragraph.

No Comments

No Known Activations

code

The neuron primarily fires on discourse-opening or connective tokens (e.g. “As,” “Sure,” “Here,” “In,” “The”) that signal the start of a response or paragraph.

No Comments

No Known Activations