INDEX

Explanations

code

np_max-act · gemini-2.0-flash

The neuron is triggered by package/import declaration tokens in source code, i.e. it detects lines that specify package and import statements.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

The neuron detects code-like text—programming language tokens, identifiers, and syntactic elements within source-code snippets.

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

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

DataManager

-0.07

 Markdown

-0.07

Plot

-0.07

Outside

-0.07

 Falk

-0.07

	left

-0.06

andidate

-0.06

_method

-0.06

Dep

-0.06

POSITIVE LOGITS

ีความ

0.06

하는

0.06

 datové

0.06

lâm

0.06

oriasis

0.06

แบ

0.06

ゅ

0.06

FindBy

0.06

VAS

0.06

.setPreferredSize

0.05

Activations Density 0.543%

code

The neuron is triggered by package/import declaration tokens in source code, i.e. it detects lines that specify package and import statements.

The neuron detects code-like text—programming language tokens, identifiers, and syntactic elements within source-code snippets.

No Comments

No Known Activations