INDEX

Explanations

tolerated

np_max-act · gemini-2.0-flash

The neuron detects mentions of treatment tolerability or safety (e.g., “well tolerated,” “safe,” and related contexts).

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

.depth

-0.07

relay

-0.07

 breathe

-0.07

png

-0.06

:index

-0.06

 yüzyıl

-0.06

 oversight

-0.06

 Monroe

-0.06

 foreign

-0.06

Muj

-0.06

POSITIVE LOGITS

 vrou

0.06

-employed

0.06

 rapport

0.06

él

0.06

Ros

0.06

 Typeface

0.06

 हट

0.06

 entra

0.06

sus

0.06

Activations Density 0.016%

tolerated

The neuron detects mentions of treatment tolerability or safety (e.g., “well tolerated,” “safe,” and related contexts).

No Comments

No Known Activations