INDEX

Explanations

phrases related to specific geographical locations, particularly Sierra and Yosemite

oai_token-act-pair · gpt-3.5-turbo

mentions of geographical locations, specifically when referring to mountains or clubs associated with them

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ifier

-0.81

ifying

-0.73

umbn

-0.70

roit

-0.70

secondary

-0.69

spring

-0.69

body

-0.67

SHIP

-0.67

animous

-0.66

orescent

-0.65

POSITIVE LOGITS

 Leone

1.29

 Sierra

1.21

ierra

1.05

 Nevada

0.89

 Scotia

0.80

 Senegal

0.78

 Designs

0.77

 Alto

0.76

 Rica

0.76

 Vista

0.75

Activations Density 0.006%

phrases related to specific geographical locations, particularly Sierra and Yosemite

mentions of geographical locations, specifically when referring to mountains or clubs associated with them

No Comments

No Known Activations