INDEX

Explanations

acronyms related to various organizations or political discussions

oai_token-act-pair · gpt-3.5-turbo

references to formal organizations and their associated acronym names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 answ

-0.85

icter

-0.72

igmatic

-0.72

 guiActiveUn

-0.68

iences

-0.67

 tremend

-0.67

zers

-0.67

 flowering

-0.67

shr

-0.65

rew

-0.65

POSITIVE LOGITS

 Committees

0.73

ãĥİ

0.71

 Ø§ÙĦ

0.68

FF

0.62

 Cooperation

0.61

eous

0.61

EY

0.61

âĢ¦)

0.60

)].

0.60

 Mines

0.60

Activations Density 0.155%

acronyms related to various organizations or political discussions

references to formal organizations and their associated acronym names

No Comments

No Known Activations