INDEX

Explanations

Nonsense text

np_max-act · gemini-2.0-flash

The neuron flags tokens corresponding to the Unicode escape sequence for apostrophes (e.g. “\u2019” and its numeric components).

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

SSID

-0.07

かの

-0.07

iß

-0.06

BOX

-0.06

erm

-0.06

Egg

-0.06

 Ethiopian

-0.06

street

-0.06

 ance

-0.06

 Audio

-0.06

POSITIVE LOGITS

.setItems

0.07

(TABLE

0.07

:view

0.07

-confirm

0.06

asmine

0.06

ynos

0.06

Cv

0.06

 Individual

0.06

Serialized

0.06

kê

0.06

Activations Density 0.045%

Nonsense text

The neuron flags tokens corresponding to the Unicode escape sequence for apostrophes (e.g. “\u2019” and its numeric components).

No Comments

No Known Activations