INDEX

Explanations

medical terms related to a specific condition

oai_token-act-pair · gpt-3.5-turbo

the term "primary" related to various contexts or subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

OTO

-0.76

 Braun

-0.66

INESS

-0.66

glass

-0.65

Ir

-0.64

estern

-0.64

 Shack

-0.63

EEK

-0.63

Dick

-0.63

ARDS

-0.63

POSITIVE LOGITS

 primary

0.86

 distingu

0.83

stay

0.77

 distinguishing

0.77

 objective

0.76

yout

0.73

alysed

0.72

 outcome

0.71

 antagonist

0.71

ities

0.69

Activations Density 0.011%

medical terms related to a specific condition

the term "primary" related to various contexts or subjects

No Comments

No Known Activations