INDEX

Explanations

Scientific and physics terms

np_max-act · gemini-2.0-flash

This neuron activates on technical adjectives ending in “-tropic” (e.g., isotropic, isentropic) in fluid-dynamics 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

 mysql

-0.06

Opr

-0.06

-aff

-0.06

EXPECTED

-0.06

 cuid

-0.06

 jMenuItem

-0.06

�

-0.06

 playwright

-0.06

-bit

-0.06

Latin

-0.06

POSITIVE LOGITS

 알아

0.08

HOT

0.07

ives

0.07

머

0.07

.Generation

0.07

 특별

0.07

 मद

0.06

meth

0.06

etter

0.06

 대해서

0.06

Activations Density 0.015%

Scientific and physics terms

This neuron activates on technical adjectives ending in “-tropic” (e.g., isotropic, isentropic) in fluid-dynamics contexts.

No Comments

No Known Activations

Scientific and physics terms

This neuron activates on technical adjectives ending in “-tropic” (e.g., isotropic, isentropic) in fluid-dynamics contexts.

No Comments

No Known Activations