INDEX

Explanations

phrases related to the keyword "Panda"

oai_token-act-pair · gpt-3.5-turbo

mentions of "Panda" and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

kick

-0.68

LECT

-0.65

grading

-0.65

RAFT

-0.64

 Eisen

-0.63

 prosec

-0.63

anwhile

-0.63

redits

-0.61

tops

-0.60

CLE

-0.60

POSITIVE LOGITS

 Pand

1.40

emonium

1.32

Pand

1.10

emic

0.99

 pand

0.89

vati

0.86

anus

0.85

 Pengu

0.85

acan

0.81

agar

0.80

Activations Density 0.006%

phrases related to the keyword "Panda"

mentions of "Panda" and related terms

No Comments

No Known Activations