INDEX

Explanations

numbers

np_max-act · gemini-2.0-flash

The neuron is detecting numerical temporal expressions (e.g. “in 30 minutes,” “in three weeks,” “in a couple of hours”).

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

 засід

-0.07

 icing

-0.07

Dark

-0.07

_patterns

-0.06

下载次数

-0.06

ecera

-0.06

 finance

-0.06

 Contracts

-0.06

-associated

-0.06

iology

-0.06

POSITIVE LOGITS

 DataType

0.08

.PORT

0.07

<Data

0.06

 getData

0.06

 acceptable

0.06

ãeste

0.06

Px

0.06

 정확

0.06

 Hyundai

0.06

 Morton

0.06

Activations Density 0.012%

numbers

The neuron is detecting numerical temporal expressions (e.g. “in 30 minutes,” “in three weeks,” “in a couple of hours”).

No Comments

No Known Activations

numbers

The neuron is detecting numerical temporal expressions (e.g. “in 30 minutes,” “in three weeks,” “in a couple of hours”).

No Comments

No Known Activations