INDEX

Explanations

of

np_max-act · gemini-2.0-flash

The neuron selectively activates on numeric tokens—especially decimals and measurement/statistical values—highlighting when numerical data (e.g. “1.56,” “95%,” “0.42”) appear.

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

%;
↵

-0.06

لمات

-0.06

 Patel

-0.06

__;

-0.06

modal

-0.06

PLAY

-0.06

_and

-0.05

tsx

-0.05

Aliases

-0.05

 Damon

-0.05

POSITIVE LOGITS

)<

0.08

).[

0.07

]=

0.07

 outing

0.07

'$

0.07

ในส

0.07

 princip

0.07

)=='

0.07

辞

0.07

)(((

0.07

Activations Density 0.147%

of

The neuron selectively activates on numeric tokens—especially decimals and measurement/statistical values—highlighting when numerical data (e.g. “1.56,” “95%,” “0.42”) appear.

No Comments

No Known Activations

of

The neuron selectively activates on numeric tokens—especially decimals and measurement/statistical values—highlighting when numerical data (e.g. “1.56,” “95%,” “0.42”) appear.

No Comments

No Known Activations