INDEX

Explanations

abbreviations/acronyms related to university and medical fields

oai_token-act-pair · gpt-3.5-turbo

references to the term "PU" and its various contexts, likely associated with entities or organizations

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

*/(

-0.75

Catalog

-0.68

ãĤ¼ãĤ¦ãĤ¹

-0.68

ãĥ¼ãĥĨãĤ£

-0.68

CBI

-0.67

shire

-0.66

 McDonnell

-0.66

 Orbital

-0.65

 Presbyter

-0.65

 Divide

-0.65

POSITIVE LOGITS

cci

1.19

pload

1.14

 Klux

1.01

amins

1.01

arez

0.99

awei

0.98

pport

0.98

itton

0.98

isine

0.97

agra

0.96

Activations Density 0.019%

abbreviations/acronyms related to university and medical fields

references to the term "PU" and its various contexts, likely associated with entities or organizations

No Comments

No Known Activations