INDEX

Explanations

addition

np_max-act · gemini-2.0-flash

The neuron is detecting discourse-level transition words (e.g., “Additionally,” “Furthermore,” “Finally”).

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

盗

-0.07

�

-0.06

 continent

-0.06

BX

-0.06

Remarks

-0.06

ン�

-0.06

$email

-0.06

ハ

-0.06

収

-0.06

วย

-0.06

POSITIVE LOGITS

-Л

0.06

},

0.06

 operatives

0.06

 arrays

0.06

 echoes

0.06

endDate

0.06

 Louise

0.06

 üzerinde

0.06

.splitext

0.06

}},

0.06

Activations Density 0.111%

addition

The neuron is detecting discourse-level transition words (e.g., “Additionally,” “Furthermore,” “Finally”).

No Comments

No Known Activations