INDEX

Explanations

locations or directions

oai_token-act-pair · gpt-3.5-turbo

references to geographical locations or specific places

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

NFL

-0.79

LGBT

-0.75

politics

-0.73

Netflix

-0.72

 emoji

-0.71

HTML

-0.70

ilogy

-0.70

FBI

-0.68

 certs

-0.67

NBC

-0.66

POSITIVE LOGITS

 mountainous

1.10

 caves

1.07

 woodland

1.05

Mt

1.02

 Canyon

1.02

 woods

1.00

 hills

0.99

 farmland

0.98

 cliffs

0.95

 Mount

0.95

Activations Density 0.326%

locations or directions

references to geographical locations or specific places

No Comments

No Known Activations