INDEX

Explanations

pone

np_max-act · gemini-2.0-flash

The neuron strongly fires on numeric tokens—especially decimal numbers and citation reference numbers.

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

 ruined

-0.07

--------↵

-0.07

 mindset

-0.06

 nhập

-0.06

Response

-0.06

 STATES

-0.06

 payments

-0.06

ρας

-0.06

TSA

-0.06

 bonuses

-0.06

POSITIVE LOGITS

/cms

0.07

autom

0.06

ократи

0.06

olan

0.06

 Jamaica

0.06

 Mortgage

0.06

USC

0.06

LE

0.06

 Cassandra

0.06

.Uint

0.06

Activations Density 0.011%

pone

The neuron strongly fires on numeric tokens—especially decimal numbers and citation reference numbers.

No Comments

No Known Activations