INDEX

Explanations

code and configurations

np_max-act · gemini-2.0-flash

The neuron fires on snippets of technical documentation that describe numeric thresholds or limits (e.g. “if the candidate file is 1% smaller than the next file’s size, then include…”).

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

会

-0.08

anta

-0.08

DB

-0.07

五

-0.07

眾

-0.06

会

-0.06

 '').

-0.06

Fol

-0.06

veillance

-0.06

 قتل

-0.06

POSITIVE LOGITS

 оборуд

0.07

 intric

0.06

ienie

0.06

_sy

0.06

mk

0.06

 पहच

0.06

_combine

0.06

 перер

0.06

.Abstract

0.06

_reverse

0.06

Activations Density 0.025%

code and configurations

The neuron fires on snippets of technical documentation that describe numeric thresholds or limits (e.g. “if the candidate file is 1% smaller than the next file’s size, then include…”).

No Comments

No Known Activations

code and configurations

The neuron fires on snippets of technical documentation that describe numeric thresholds or limits (e.g. “if the candidate file is 1% smaller than the next file’s size, then include…”).

No Comments

No Known Activations