INDEX

Explanations

phrases related to specific national parks or locations

oai_token-act-pair · gpt-3.5-turbo

references to national parks or national sites

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

ullivan

-0.84

enger

-0.78

areth

-0.73

roxy

-0.72

utenberg

-0.72

elf

-0.67

raq

-0.66

redd

-0.65

resp

-0.65

ements

-0.65

POSITIVE LOGITS

 Geographic

1.09

 Laboratory

0.89

 Guard

0.89

 Treasure

0.88

ity

0.84

 Monument

0.84

 Laboratories

0.83

 Park

0.83

 Recreation

0.81

 Institutes

0.79

Activations Density 0.024%

phrases related to specific national parks or locations

references to national parks or national sites

No Comments

No Known Activations