INDEX

Explanations

text related to explanations, justifications, or the underlying logic behind actions or decisions

oai_token-act-pair · gpt-3.5-turbo

phrases related to reasoning and justification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

adal

-0.72

vette

-0.68

onies

-0.67

semble

-0.67

uckle

-0.66

 stocking

-0.65

hold

-0.65

 national

-0.64

borg

-0.64

vas

-0.63

POSITIVE LOGITS

 reasoning

1.19

 rationale

1.01

DragonMagazine

0.96

why

0.95

SourceFile

0.95

 argument

0.88

 justification

0.84

 arguments

0.80

 excuse

0.79

WHY

0.79

Activations Density 0.007%

text related to explanations, justifications, or the underlying logic behind actions or decisions

phrases related to reasoning and justification

No Comments

No Known Activations