INDEX

Explanations

replace

np_max-act · gemini-2.0-flash

The neuron fires on code‐style identifiers and programming tokens, i.e. fragments of source code or API names.

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 VERSION

-0.07

 xpos

-0.07

ouples

-0.07

 pours

-0.07

 Wrapper

-0.06

 Зав

-0.06

 indirectly

-0.06

ocoder

-0.06

 holder

-0.06

anzeigen

-0.06

POSITIVE LOGITS

 tidal

0.07

าหล

0.07

 television

0.06

ASF

0.06

แปลง

0.06

 berk

0.06

 solids

0.06

 español

0.06

holiday

0.06

EDA

0.06

Activations Density 0.024%

replace

The neuron fires on code‐style identifiers and programming tokens, i.e. fragments of source code or API names.

No Comments

No Known Activations