INDEX

Explanations

size and growth

np_max-act · gemini-2.0-flash

This neuron detects numeric word-count constraint phrases (e.g. “expand more than 18 words”).

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

 Listening

-0.08

 podcasts

-0.07

_prof

-0.06

 lower

-0.06

condition

-0.06

紹介

-0.06

посеред

-0.06

ูงส

-0.06

 polymer

-0.06

细

-0.06

POSITIVE LOGITS

 perpetr

0.06

_ALIGN

0.06

zel

0.06

 Pradesh

0.06

 cread

0.06

ائد

0.06

 dici

0.06

 유지

0.06

+d

0.06

$path

0.06

Activations Density 0.003%

size and growth

This neuron detects numeric word-count constraint phrases (e.g. “expand more than 18 words”).

No Comments

No Known Activations