INDEX

Explanations

locations and their characteristics, such as cities, towns, and specific features

oai_token-act-pair · gpt-3.5-turbo

nouns indicating specific geographical locations or types of places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 knives

-0.72

 torped

-0.71

 Robots

-0.70

 refunds

-0.70

 hats

-0.68

rists

-0.67

SLI

-0.66

AUD

-0.65

docs

-0.64

ipers

-0.64

POSITIVE LOGITS

 suburb

1.42

 birthplace

1.26

 stronghold

1.20

 sprawling

1.20

 borough

1.15

 densely

1.12

 destination

1.10

 district

1.10

 city

1.10

 bustling

1.09

Activations Density 0.290%

locations and their characteristics, such as cities, towns, and specific features

nouns indicating specific geographical locations or types of places

No Comments

No Known Activations