INDEX

Explanations

terms related to architecture and architects

oai_token-act-pair · gpt-3.5-turbo

terms related to architecture and architects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

TPS

-0.76

ting

-0.74

awar

-0.72

zees

-0.70

nir

-0.70

eworthy

-0.68

othal

-0.67

zee

-0.66

cow

-0.62

ml

-0.62

POSITIVE LOGITS

itect

1.50

urally

1.50

ural

1.27

 architect

1.03

 Architecture

0.97

 architecture

0.94

 architectural

0.92

 architects

0.89

 Architect

0.85

Arch

0.81

Activations Density 0.056%

terms related to architecture and architects

terms related to architecture and architects

No Comments

No Known Activations