INDEX

Explanations

acronyms and abbreviations

oai_token-act-pair · gpt-3.5-turbo

references to specific organizations, industries, or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Columb

-0.70

raising

-0.64

 withd

-0.64

BILITIES

-0.60

 appro

-0.60

eatures

-0.59

:{

-0.59

between

-0.58

acters

-0.58

blogspot

-0.57

POSITIVE LOGITS

idia

0.82

abies

0.73

acht

0.72

phrine

0.71

ilver

0.69

onsense

0.67

zsche

0.67

ovych

0.67

oxide

0.66

ensical

0.63

Activations Density 0.416%

acronyms and abbreviations

references to specific organizations, industries, or entities

No Comments

No Known Activations