INDEX

Explanations

Negation

np_max-act · gemini-2.0-flash

The neuron fires on capitalized “Not” (and adjacent tokens like “Exists,” “Supported,” etc.) when used in exception or error‐class names to signal a negative condition.

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

 gram

-0.06

_LEN

-0.06

 arrest

-0.06

OfSize

-0.06

 OPTION

-0.06

SEN

-0.06

 brig

-0.06

[model

-0.06

.ping

-0.06

 Science

-0.06

POSITIVE LOGITS

 الكتاب

0.07

�

0.07

_hover

0.06

랫

0.06

РА

0.06

 Tommy

0.06

 кош

0.06

…………………………………………

0.06

 أبو

0.06

Tumblr

0.06

Activations Density 0.016%

Negation

The neuron fires on capitalized “Not” (and adjacent tokens like “Exists,” “Supported,” etc.) when used in exception or error‐class names to signal a negative condition.

No Comments

No Known Activations

Negation

The neuron fires on capitalized “Not” (and adjacent tokens like “Exists,” “Supported,” etc.) when used in exception or error‐class names to signal a negative condition.

No Comments

No Known Activations