INDEX

Explanations

support

np_max-act · gemini-2.0-flash

The neuron detects mentions of “support(s)” or compatibility features (i.e. when something is said to “support” or “be supported” by a client, device, or feature).

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

_generation

-0.07

给

-0.07

Democratic

-0.07

uity

-0.07

arem

-0.07

 employment

-0.07

 život

-0.06

_CARD

-0.06

 fiery

-0.06

(\

-0.06

POSITIVE LOGITS

飞

0.07

 phân

0.07

TEL

0.06

imiter

0.06

](↵

0.06

ٹ

0.06

 này

0.06

vis

0.06

iram

0.06

 Warning

0.06

Activations Density 0.026%

support

The neuron detects mentions of “support(s)” or compatibility features (i.e. when something is said to “support” or “be supported” by a client, device, or feature).

No Comments

No Known Activations