INDEX

Explanations

news or updates related to relationships, interviews, or projects involving public figures

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

 pand

-0.79

adal

-0.69

bard

-0.69

pan

-0.68

milo

-0.66

ĸļ

-0.66

heid

-0.65

 theorem

-0.64

eg

-0.63

gaard

-0.62

POSITIVE LOGITS

 exclusively

0.95

TMZ

0.88

 leaks

0.82

 specs

0.82

 revealing

0.80

 confirming

0.80

 confirms

0.80

 IMAGES

0.79

 leaked

0.79

 details

0.79

Activations Density 0.211%

news or updates related to relationships, interviews, or projects involving public figures

No Comments

No Known Activations