INDEX

Explanations

medical terms or references related to diabetes

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "di" across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

manship

-0.77

 Guarant

-0.74

essee

-0.74

EVs

-0.67

 suspense

-0.65

 Chill

-0.65

 Scenes

-0.65

*/(

-0.63

ï¸ı

-0.62

 Charlotte

-0.62

POSITIVE LOGITS

plom

1.38

urnal

1.15

ablo

1.14

di

1.08

Di

1.08

abetes

1.06

agonal

1.05

ocese

1.03

ploma

1.03

orno

0.97

Activations Density 0.005%

medical terms or references related to diabetes

occurrences of the substring "di" across various contexts

No Comments

No Known Activations