INDEX

Explanations

adjectives denoting strong positive emotions

oai_token-act-pair · gpt-3.5-turbo

expressions of amazement or admiration regarding various subjects

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

erent

-0.88

agle

-0.74

eeper

-0.73

eret

-0.71

odox

-0.71

eper

-0.71

enhagen

-0.69

veland

-0.68

 stale

-0.67

istant

-0.66

POSITIVE LOGITS

 ecstasy

0.76

 pandemonium

0.74

 Beaut

0.73

 proportions

0.73

 unbeliev

0.73

Beaut

0.70

 Beasts

0.68

 feats

0.68

 è£ıç

0.67

Merit

0.66

Activations Density 0.587%

adjectives denoting strong positive emotions

expressions of amazement or admiration regarding various subjects

No Comments

No Known Activations