INDEX

Explanations

phrases related to historical events and disasters

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "in" within various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wcsstore

-0.76

leans

-0.68

 ingred

-0.68

rawdownloadcloneembedreportprint

-0.68

minist

-0.67

natureconservancy

-0.66

issance

-0.65

gow

-0.65

BILITIES

-0.65

dit

-0.63

POSITIVE LOGITS

 conjunction

1.02

effect

0.91

0.89

 Greece

0.89

 Yugoslavia

0.89

effic

0.88

 favor

0.88

 Ukraine

0.88

 Cambodia

0.87

Activations Density 0.251%

phrases related to historical events and disasters

occurrences of the word "in" within various contexts

No Comments

No Known Activations