INDEX

Explanations

acronyms and abbreviations

oai_token-act-pair · gpt-3.5-turbo

terms related to acronyms and abbreviations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Thoughts

-0.70

igators

-0.69

ebook

-0.65

 Huang

-0.64

Peb

-0.63

 Diver

-0.63

 Blasio

-0.63

taboola

-0.63

 Exper

-0.63

 Emin

-0.62

POSITIVE LOGITS

viation

1.31

 acronym

1.12

 abbre

1.12

 abbrevi

1.10

 acron

0.90

 shortened

0.90

 notation

0.88

 prefix

0.84

 shorthand

0.83

notation

0.78

Activations Density 0.022%

acronyms and abbreviations

terms related to acronyms and abbreviations

No Comments

No Known Activations