INDEX

Explanations

positive comments or expressions in social media or communication

oai_token-act-pair · gpt-3.5-turbo

emojis and emoticons expressing positive or neutral emotions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

lings

-0.92

ling

-0.74

 displacement

-0.73

ciples

-0.72

©¶æ

-0.69

 disputed

-0.69

eport

-0.68

yer

-0.67

 grapp

-0.66

 subsequ

-0.66

POSITIVE LOGITS

:)

1.05

:-)

1.03

;)

0.95

:(

0.90

 Anyway

0.85

 ðŁĻĤ

0.85

 Sorry

0.83

#$

0.82

[/

0.82

 Thank

0.81

Activations Density 0.012%

positive comments or expressions in social media or communication

emojis and emoticons expressing positive or neutral emotions

No Comments

No Known Activations