INDEX

Explanations

Identifiers and version numbers

np_max-act · gemini-2.0-flash

This neuron fires on tokens that are part of product or model version identifiers (e.g., “T-2,” “2.1,” “G4c,” “X8,” “.B” variants).

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

 dungeon

-0.07

 tactical

-0.07

폴

-0.06

 subscription

-0.06

.sensor

-0.06

.Var

-0.06

 거의

-0.06

 менее

-0.06

 드립니다

-0.06

 correct

-0.06

POSITIVE LOGITS

(TYPE

0.08

 Evidence

0.07

ometer

0.06

ційних

0.06

Doctrine

0.06

PAR

0.06

 PACK

0.06

'?'

0.06

ometers

0.06

 incontr

0.06

Activations Density 0.060%

Identifiers and version numbers

This neuron fires on tokens that are part of product or model version identifiers (e.g., “T-2,” “2.1,” “G4c,” “X8,” “.B” variants).

No Comments

No Known Activations