INDEX

Explanations

phrases or sentences where an action is being initiated or discussed

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate actions or responses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

inations

-0.71

istant

-0.68

SPONSORED

-0.67

adoes

-0.63

tan

-0.62

adal

-0.61

isable

-0.60

Merc

-0.60

din

-0.60

Silver

-0.59

POSITIVE LOGITS

 virtue

1.16

products

1.01

 placing

0.95

laws

0.95

 eliminating

0.94

 adding

0.92

 leaps

0.91

 default

0.89

 putting

0.89

 introducing

0.88

Activations Density 0.084%

phrases or sentences where an action is being initiated or discussed

phrases that indicate actions or responses

No Comments

No Known Activations