INDEX

Explanations

numbers

np_max-act · gemini-2.0-flash

The neuron fires on code-like text—blocks of programming language syntax (keywords, identifiers, braces, XML tags, etc.) rather than natural language.

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

;s

-0.06

.loader

-0.06

buf

-0.06

os

-0.06

ارش

-0.06

ξεις

-0.06

VM

-0.06

nutím

-0.06

upert

-0.06

Sale

-0.06

POSITIVE LOGITS

liği

0.07

<Void

0.06

 dům

0.06

DEV

0.06

.\

0.06

 hangi

0.06

.deg

0.06

rua

0.06

vũ

0.06

*)&

0.06

Activations Density 0.114%

numbers

The neuron fires on code-like text—blocks of programming language syntax (keywords, identifiers, braces, XML tags, etc.) rather than natural language.

No Comments

No Known Activations