INDEX

Explanations

phrases or sentences that prompt the reader to take an action

oai_token-act-pair · gpt-3.5-turbo

instances of the word "the" and similar contextual phrases

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

iatus

-0.78

tsy

-0.71

ufact

-0.67

warts

-0.65

enegger

-0.64

 Constantin

-0.63

llor

-0.61

 Baghd

-0.61

daughter

-0.60

emetery

-0.60

POSITIVE LOGITS

 levers

0.88

 sparing

0.86

 hashtag

0.82

 borrowed

0.78

 pseudonym

0.78

 newfound

0.77

 tools

0.77

 tactic

0.76

 metaphor

0.73

 analogy

0.72

Activations Density 0.214%

phrases or sentences that prompt the reader to take an action

instances of the word "the" and similar contextual phrases

No Comments

No Known Activations