INDEX

Explanations

terms related to science fiction

oai_token-act-pair · gpt-3.5-turbo

references to science fiction elements and themes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

lain

-1.00

theless

-0.91

xual

-0.83

 Mour

-0.81

LIA

-0.75

hof

-0.75

upon

-0.74

flies

-0.74

 largeDownload

-0.74

LOAD

-0.73

POSITIVE LOGITS

sci

1.14

 fiction

0.86

adelphia

0.83

 geek

0.75

ECA

0.70

fi

0.70

posium

0.69

 futuristic

0.68

 nerd

0.68

 shorts

0.67

Activations Density 0.007%

terms related to science fiction

references to science fiction elements and themes

No Comments

No Known Activations