INDEX

Explanations

locations based on geographical locations along with a business or professional context

oai_token-act-pair · gpt-3.5-turbo

references to locations paired with descriptors related to individuals or 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

 Replay

-0.84

 elim

-0.74

 replay

-0.73

 sacrific

-0.73

 headlights

-0.72

 criterion

-0.70

 revers

-0.68

 recess

-0.68

SPI

-0.68

 stripe

-0.66

POSITIVE LOGITS

based

1.81

area

1.60

centric

1.29

raised

1.27

born

1.27

trained

1.27

native

1.26

focused

1.23

educated

1.20

themed

1.19

Activations Density 0.031%

locations based on geographical locations along with a business or professional context

references to locations paired with descriptors related to individuals or entities

No Comments

No Known Activations