INDEX

Explanations

mentions of historical events, particularly World War II related terms

oai_token-act-pair · gpt-3.5-turbo

references to World War II and related historical contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Thumbnails

-0.83

ursday

-0.75

haw

-0.72

erd

-0.71

Wiz

-0.69

ciating

-0.67

Discussion

-0.67

ailable

-0.66

BOOK

-0.65

copy

-0.64

POSITIVE LOGITS

era

1.14

Era

0.95

 eras

0.90

 battlefield

0.86

 trenches

0.83

 Allied

0.83

 period

0.80

 occupation

0.79

GI

0.79

 onwards

0.78

Activations Density 0.075%

mentions of historical events, particularly World War II related terms

references to World War II and related historical contexts

No Comments

No Known Activations