INDEX

Explanations

proper names and detailed information related to academia, researchers, and professionals

oai_token-act-pair · gpt-3.5-turbo

key figures or names associated with significant research or actions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Interstitial

-0.86

 Confederate

-0.85

ansas

-0.78

 diaper

-0.74

 pickup

-0.72

 vigilante

-0.70

 HERO

-0.70

 pageant

-0.70

 spitting

-0.70

 bumper

-0.70

POSITIVE LOGITS

ijn

1.11

ijk

1.11

et

1.09

sson

0.97

 Zhang

0.95

 Huang

0.91

Xu

0.91

essler

0.91

 argues

0.90

Rao

0.90

Activations Density 0.361%

proper names and detailed information related to academia, researchers, and professionals

key figures or names associated with significant research or actions

No Comments

No Known Activations