INDEX

Explanations

hyphen "-"

np_max-act · gemini-2.0-flash

This neuron detects mentions of GPT model version identifiers (e.g. “GPT-3,” “GPT-4”).

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

crow

-0.07

 whistle

-0.07

 Sand

-0.07

Sak

-0.07

 Kumar

-0.06

Amb

-0.06

 eagerly

-0.06

 cents

-0.06

Sw

-0.06

 racist

-0.06

POSITIVE LOGITS

yı

0.07

 istediğiniz

0.06

его

0.06

/photo

0.06

_associ

0.06

Pitch

0.06

@Before

0.06

۲۰۲

0.06

ingular

0.06

……↵↵

0.06

Activations Density 0.003%

hyphen "-"

This neuron detects mentions of GPT model version identifiers (e.g. “GPT-3,” “GPT-4”).

No Comments

No Known Activations

hyphen "-"

This neuron detects mentions of GPT model version identifiers (e.g. “GPT-3,” “GPT-4”).

No Comments

No Known Activations