INDEX

Explanations

organizations or concepts related to healthcare or caregiving

oai_token-act-pair · gpt-3.5-turbo

references to care-related topics or entities

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

ãĥĥãĥī

-0.87

 Chomsky

-0.65

 extrad

-0.63

lap

-0.61

integer

-0.61

 poppy

-0.60

 Kenyan

-0.59

nav

-0.59

âĶĢâĶĢâĶĢâĶĢ

-0.58

 phyl

-0.57

POSITIVE LOGITS

taker

1.09

er

1.05

care

0.97

ndra

0.90

lli

0.89

llan

0.88

giving

0.85

getic

0.81

tta

0.79

atism

0.78

Activations Density 0.013%

organizations or concepts related to healthcare or caregiving

references to care-related topics or entities

No Comments

No Known Activations