INDEX

Explanations

academic papers

np_max-act · gemini-2.0-flash

The neuron activates on words signalling the paper’s stated goals or objectives (e.g. “aim,” “purpose,” “intend”) in academic introductions.

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

 Workers

-0.07

,p

-0.06

PA

-0.06

Word

-0.06

[h

-0.06

Mos

-0.06

[,

-0.06

Ja

-0.06

ﾘ

-0.06

ево

-0.06

POSITIVE LOGITS

登

0.07

.isFile

0.07

	void

0.07

sf

0.07

zip

0.07

 Curve

0.06

 beast

0.06

sig

0.06

ellery

0.06

rowave

0.06

Activations Density 0.020%

academic papers

The neuron activates on words signalling the paper’s stated goals or objectives (e.g. “aim,” “purpose,” “intend”) in academic introductions.

No Comments

No Known Activations

academic papers

The neuron activates on words signalling the paper’s stated goals or objectives (e.g. “aim,” “purpose,” “intend”) in academic introductions.

No Comments

No Known Activations