INDEX

Explanations

specific pairs of names that are frequently mentioned together

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate connections or comparisons between entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ancial

-0.86

incial

-0.80

doctor

-0.73

Asian

-0.70

cms

-0.70

Location

-0.67

ometime

-0.66

istrate

-0.65

sylv

-0.65

ĻĤ

-0.64

POSITIVE LOGITS

 Sanchez

1.19

 Daniels

1.18

 Adams

1.16

 Perez

1.16

 Lopez

1.15

 Martinez

1.15

 Moore

1.15

 McCarthy

1.14

 Stein

1.14

 Johnson

1.14

Activations Density 0.184%

specific pairs of names that are frequently mentioned together

phrases that indicate connections or comparisons between entities

No Comments

No Known Activations