INDEX

Explanations

phrases related to proper nouns, such as names of people, places, organizations, and titles

oai_token-act-pair · gpt-3.5-turbo

parentheses and brackets, indicating references and citations

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

 joints

-0.60

adm

-0.60

 inspir

-0.59

 retina

-0.58

 plagiar

-0.57

 royalty

-0.57

 seams

-0.56

 renown

-0.56

 clipboard

-0.56

 commissions

-0.56

POSITIVE LOGITS

RELATED

1.02

Reuters

0.97

Reporting

0.97

PHOTOS

0.88

Read

0.86

np

0.85

CNN

0.85

WATCH

0.84

ABC

0.83

Polit

0.83

Activations Density 0.180%

phrases related to proper nouns, such as names of people, places, organizations, and titles

parentheses and brackets, indicating references and citations

No Comments

No Known Activations