INDEX

Explanations

positive and enthusiastic exclamations

oai_token-act-pair · gpt-3.5-turbo

punctuation marks and symbols, particularly those representing positive or negative sentiments

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

 skelet

-0.64

States

-0.64

REE

-0.62

 Continued

-0.61

 Mour

-0.58

FAR

-0.57

ľ

-0.57

ãģ®å

-0.57

NCT

-0.56

conom

-0.56

POSITIVE LOGITS

ooters

1.10

iverpool

0.85

atform

0.73

itaire

0.72

ammu

0.71

ooter

0.71

wark

0.69

oice

0.68

iewicz

0.67

 which

0.66

Activations Density 0.031%

positive and enthusiastic exclamations

punctuation marks and symbols, particularly those representing positive or negative sentiments

No Comments

No Known Activations