INDEX

Explanations

positively

np_max-act · gemini-2.0-flash

The neuron flags adverbs and phrases that denote the direction of statistical associations (e.g. “positively” or “negatively” correlated/associated).

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

Wilson

-0.07

 clumsy

-0.07

 drastic

-0.07

 vielen

-0.07

Take

-0.07

/team

-0.07

 milk

-0.07

Paper

-0.06

.Depth

-0.06

Ten

-0.06

POSITIVE LOGITS

 bisa

0.07

(::

0.06

 rails

0.06

.Mapper

0.06

("{\"

0.06

(&$

0.06

 _______,

0.06

##↵↵

0.06

'['

0.06

.DeserializeObject

0.06

Activations Density 0.011%

positively

The neuron flags adverbs and phrases that denote the direction of statistical associations (e.g. “positively” or “negatively” correlated/associated).

No Comments

No Known Activations

positively

The neuron flags adverbs and phrases that denote the direction of statistical associations (e.g. “positively” or “negatively” correlated/associated).

No Comments

No Known Activations