INDEX

Explanations

Code/Formulas

np_max-act · gemini-2.0-flash

The neuron primarily responds to the various numerical indices and structural tokens used in the document’s parsing metadata (e.g. the numbers labeling words, tree-node positions, and bracketed annotation slots).

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

ustom

-0.08

.Build

-0.06

 Lets

-0.06

其实

-0.06

 tenant

-0.06

istrib

-0.06

_HINT

-0.06

 रहत

-0.06

 صند

-0.06

 годы

-0.06

POSITIVE LOGITS

ايا

0.07

 Львів

0.06

uguay

0.06

 okum

0.06

*/

0.06

sat

0.06

 silah

0.06

يلا

0.06

.domain

0.06

 "-",

0.06

Activations Density 0.002%

Code/Formulas

The neuron primarily responds to the various numerical indices and structural tokens used in the document’s parsing metadata (e.g. the numbers labeling words, tree-node positions, and bracketed annotation slots).

No Comments

No Known Activations