INDEX

Explanations

words related to the human nervous system

oai_token-act-pair · gpt-3.5-turbo

terms related to psychological or medical conditions

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

 liking

-0.64

 basis

-0.63

 Collider

-0.63

 Byrd

-0.62

 Wizards

-0.60

 shin

-0.59

 Keystone

-0.59

$$$$

-0.57

 Dayton

-0.57

oglu

-0.56

POSITIVE LOGITS

athetic

1.03

abus

0.95

esthesia

0.94

etry

0.92

agog

0.89

hetic

0.89

olic

0.89

getic

0.88

achus

0.87

etrical

0.86

Activations Density 0.075%

words related to the human nervous system

terms related to psychological or medical conditions

No Comments

No Known Activations