INDEX

Explanations

references to healing, including healers, wound healing, and the healing process

oai_token-act-pair · gpt-3.5-turbo

terms associated with healing and recovery processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

yip

-0.70

legal

-0.68

urity

-0.67

SpaceEngineers

-0.66

azeera

-0.64

*/(

-0.62

atography

-0.58

Sop

-0.57

ozy

-0.57

rance

-0.57

POSITIVE LOGITS

 wounds

1.08

 heal

1.01

 healed

1.00

 healing

0.97

 heals

0.97

 Healing

0.80

ilitating

0.79

erest

0.74

 ashore

0.74

ened

0.71

Activations Density 0.058%

references to healing, including healers, wound healing, and the healing process

terms associated with healing and recovery processes

No Comments

No Known Activations