INDEX

Explanations

changes

np_max-act · gemini-2.0-flash

The neuron activates on tokens within the assistant’s “breakdown of changes” or edit-explanation section (e.g. “Here,” “breakdown,” numbered list markers, and descriptions of edits).

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

 mutable

-0.07

"%

-0.07

 fruits

-0.06

.message

-0.06

Common

-0.06

'%

-0.06

 barn

-0.06

PyObject

-0.06

(write

-0.06

 Buff

-0.06

POSITIVE LOGITS

quette

0.07

ブラ

0.07

 elles

0.07

 retrospective

0.07

pecia

0.06

kün

0.06

 đông

0.06

گری

0.06

張

0.06

 agitation

0.06

Activations Density 0.027%

changes

The neuron activates on tokens within the assistant’s “breakdown of changes” or edit-explanation section (e.g. “Here,” “breakdown,” numbered list markers, and descriptions of edits).

No Comments

No Known Activations