INDEX

Explanations

code/technical data

np_max-act · gemini-2.0-flash

The neuron predominantly activates on numeric literal tokens (especially floating‐point constants) in code.

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

げ

-0.07

_rpc

-0.07

 Serialization

-0.07

fro

-0.07

_multiply

-0.07

fw

-0.07

.UltraWin

-0.06

ifica

-0.06

_type

-0.06

wx

-0.06

POSITIVE LOGITS

ВО

0.07

black

0.06

odem

0.06

Ос

0.06

')(

0.06

endency

0.06

DNC

0.06

 However

0.06

ーー

0.06

Activations Density 0.032%

code/technical data

The neuron predominantly activates on numeric literal tokens (especially floating‐point constants) in code.

No Comments

No Known Activations