INDEX

Explanations

code

np_max-act · gemini-2.0-flash

The neuron is essentially picking out numeric tokens (digits or numbers) in the text.

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

ama

-0.06

isclosed

-0.06

.buffer

-0.06

 стратег

-0.06

.kafka

-0.06

XF

-0.06

лению

-0.06

[{

-0.05

 kittens

-0.05

_TIMEOUT

-0.05

POSITIVE LOGITS

 yaptığı

0.07

 noen

0.07

 تغییر

0.07

 AndAlso

0.07

pons

0.07

 virt

0.07

 ettiği

0.07

crets

0.06

structural

0.06

ایت

0.06

Activations Density 0.051%

code

The neuron is essentially picking out numeric tokens (digits or numbers) in the text.

No Comments

No Known Activations