INDEX

Explanations

abbreviations associated with different types of services or organizations

oai_token-act-pair · gpt-3.5-turbo

references to formal organizations or acts, particularly those involving abbreviations or acronyms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Ļ½

-0.73

spo

-0.69

oho

-0.69

ãĥ´ãĤ¡

-0.68

(&

-0.66

((

-0.65

ãĥĦ

-0.65

ãĥ¥

-0.62

ãĥ´

-0.62

:(

-0.61

POSITIVE LOGITS

/"

0.62

 Certification

0.62

 Complex

0.60

 Committees

0.59

ventures

0.58

udence

0.58

 Syndicate

0.58

Ltd

0.57

*.

0.55

system

0.54

Activations Density 0.114%

abbreviations associated with different types of services or organizations

references to formal organizations or acts, particularly those involving abbreviations or acronyms

No Comments

No Known Activations