INDEX

Explanations

phrases related to research, investigations, reports, and inquiries

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the need for further investigation or research

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ghazi

-0.76

avis

-0.68

Cele

-0.62

Dest

-0.60

 Corpus

-0.58

 cruc

-0.58

iverse

-0.58

 Himself

-0.57

 Porn

-0.57

 Deals

-0.57

POSITIVE LOGITS

 ensued

1.03

 emerges

0.84

 awaits

0.80

 aside

0.77

 prevailed

0.76

 sprang

0.76

 emerged

0.74

 analysis

0.74

 arose

0.73

 develops

0.72

Activations Density 0.461%

phrases related to research, investigations, reports, and inquiries

phrases indicating the need for further investigation or research

No Comments

No Known Activations