INDEX

Explanations

code examples

np_max-act · gemini-2.0-flash

The neuron fires on instructional cue phrases that introduce code snippets—e.g. “Add this code,” “the following code,” or “add this line.”

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

iphery

-0.07

уже

-0.06

пис

-0.06

	tile

-0.06

amples

-0.06

illage

-0.06

 завод

-0.06

	padding

-0.06

perf

-0.06

ponent

-0.06

POSITIVE LOGITS

 громадян

0.07

SE

0.07

secret

0.07

NEC

0.06

aer

0.06

Enc

0.06

cv

0.06

 sane

0.06

�

0.06

{}{↵

0.06

Activations Density 0.023%

code examples

The neuron fires on instructional cue phrases that introduce code snippets—e.g. “Add this code,” “the following code,” or “add this line.”

No Comments

No Known Activations

code examples

The neuron fires on instructional cue phrases that introduce code snippets—e.g. “Add this code,” “the following code,” or “add this line.”

No Comments

No Known Activations