INDEX

Explanations

references to geographical regions or locations

oai_token-act-pair · gpt-3.5-turbo

references to geographical regions, specifically focused on continental and mainland distinctions

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

IFrame

Link

Negative Logits

HUD

-0.78

endo

-0.71

atis

-0.69

oga

-0.69

 Survivors

-0.68

verbs

-0.68

ended

-0.68

tu

-0.68

 Skydragon

-0.67

SA

-0.66

POSITIVE LOGITS

ciating

0.85

llan

0.80

olate

0.74

tesy

0.68

continental

0.68

 shelves

0.68

 drift

0.68

 shelf

0.66

inian

0.66

theless

0.64

Activations Density 0.100%

references to geographical regions or locations

references to geographical regions, specifically focused on continental and mainland distinctions

No Known Activations