INDEX

Explanations

abbreviations related to academic or professional titles

oai_token-act-pair · gpt-3.5-turbo

abbreviations or initials that often correlate with awards or recognitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 toile

-0.79

 questioning

-0.63

 succession

-0.60

 scripting

-0.60

 omission

-0.59

 document

-0.59

 forth

-0.59

humans

-0.58

 Notting

-0.58

 stewards

-0.58

POSITIVE LOGITS

gage

1.00

ufact

0.91

icio

0.84

udd

0.84

waukee

0.82

achine

0.80

ICAN

0.78

\-

0.75

ascar

0.75

ains

0.74

Activations Density 0.042%

abbreviations related to academic or professional titles

abbreviations or initials that often correlate with awards or recognitions

No Comments

No Known Activations