INDEX

Explanations

code

np_max-act · gemini-2.0-flash

The neuron activates on tokens naming common “output” or “write”–type functions (e.g. …_write, printf/_fprintf, rb_ary_push, etc.), i.e. it flags function calls that perform writing or printing.

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

شته

-0.07

_ratings

-0.06

 guidelines

-0.06

_reviews

-0.06

deep

-0.06

多

-0.06

ували

-0.06

 based

-0.06

 sets

-0.06

ابي

-0.06

POSITIVE LOGITS

 gehen

0.08

 presentViewController

0.07

qa

0.07

reon

0.07

kes

0.06

 forensic

0.06

])/

0.06

.appendChild

0.06

 etme

0.06

/ts

0.06

Activations Density 0.028%

code

The neuron activates on tokens naming common “output” or “write”–type functions (e.g. …_write, printf/_fprintf, rb_ary_push, etc.), i.e. it flags function calls that perform writing or printing.

No Comments

No Known Activations

code

The neuron activates on tokens naming common “output” or “write”–type functions (e.g. …_write, printf/_fprintf, rb_ary_push, etc.), i.e. it flags function calls that perform writing or printing.

No Comments

No Known Activations