INDEX

Explanations

technical terms and acronyms related to various fields

oai_token-act-pair · gpt-3.5-turbo

occurrences of parentheses and brackets, which often denote additional information or clarifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

quartered

-0.68

ttes

-0.66

atu

-0.65

nikov

-0.63

ilus

-0.62

otide

-0.62

lie

-0.62

isky

-0.60

Tus

-0.60

angelo

-0.60

POSITIVE LOGITS

ASY

0.91

NET

0.91

ITION

0.81

RD

0.81

COM

0.79

RANT

0.79

DEN

0.79

HAM

0.78

POL

0.78

FD

0.77

Activations Density 0.401%

technical terms and acronyms related to various fields

occurrences of parentheses and brackets, which often denote additional information or clarifications

No Comments

No Known Activations