INDEX

Explanations

words related to wounds and medical treatments

oai_token-act-pair · gpt-3.5-turbo

references to injuries and wounds

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

soDeliveryDate

-0.92

*/(

-0.73

ructose

-0.71

racuse

-0.69

Balance

-0.69

 GEAR

-0.68

bnb

-0.68

oaded

-0.67

inyl

-0.67

VW

-0.66

POSITIVE LOGITS

 wounds

1.65

 wound

1.33

 inflicted

1.32

 healed

1.30

 heals

1.19

 scars

1.19

 heal

1.18

 healing

1.09

 sore

0.99

 bleeding

0.98

Activations Density 0.094%

words related to wounds and medical treatments

references to injuries and wounds

No Comments

No Known Activations