INDEX

Explanations

phrases related to specific locations or events, particularly with N.C. or N.Y. abbreviations

oai_token-act-pair · gpt-3.5-turbo

specific geographical identifiers and references to locations

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

Not in Any Lists

No Comments

Negative Logits

 Situation

-0.70

DragonMagazine

-0.67

uddin

-0.63

-0.60

ufact

-0.60

idis

-0.59

emort

-0.59

 roadside

-0.59

 Tsarnaev

-0.59

 Abedin

-0.58

POSITIVE LOGITS

sylvania

0.84

.,

0.73

 Tenn

0.68

astern

0.68

essee

0.67

achus

0.67

emia

0.66

isco

0.66

ansas

0.66

 Calif

0.66

Activations Density 0.088%

phrases related to specific locations or events, particularly with N.C. or N.Y. abbreviations

specific geographical identifiers and references to locations

No Comments

No Known Activations