INDEX

Explanations

Code

np_max-act · gemini-2.0-flash

The neuron fires on contiguous runs of programming‐language tokens—i.e. lines of code or code‐like syntax.

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

 Rossi

-0.07

stří

-0.07

mış

-0.07

ناء

-0.06

 باست

-0.06

 NSDate

-0.06

EMON

-0.06

arcer

-0.06

 compte

-0.06

 пош

-0.06

POSITIVE LOGITS

wag

0.07

sor

0.06

 stole

0.06

'=>$

0.06

 stereotype

0.06

nob

0.06

=current

0.06

elo

0.06

 convicted

0.06

 leopard

0.06

Activations Density 0.016%

Code

The neuron fires on contiguous runs of programming‐language tokens—i.e. lines of code or code‐like syntax.

No Comments

No Known Activations