INDEX

Explanations

mentions of the term "Jar" followed by a number, potentially referring to a specific context or entity

oai_token-act-pair · gpt-3.5-turbo

references to "jar."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Samoa

-0.74

ĨĴ

-0.72

EPA

-0.69

ãģį

-0.68

Premium

-0.67

 Mutual

-0.65

 Hawaiian

-0.65

Ü

-0.63

 Canterbury

-0.62

 Ascend

-0.61

POSITIVE LOGITS

Jar

1.11

assic

0.98

olla

0.96

Jar

0.95

oche

0.93

uda

0.91

oon

0.88

rod

0.85

abul

0.85

ney

0.83

Activations Density 0.025%

mentions of the term "Jar" followed by a number, potentially referring to a specific context or entity

references to "jar."

No Comments

No Known Activations