INDEX

Explanations

words related to chemical substances, particularly glycerin

oai_token-act-pair · gpt-3.5-turbo

words related to bodily fluids and medical conditions

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

Negative Logits

 Kund

-0.73

 Skydragon

-0.70

TAM

-0.64

 Principles

-0.63

 judgement

-0.60

 spare

-0.60

 reins

-0.60

Lak

-0.59

¥µ

-0.58

 departments

-0.58

POSITIVE LOGITS

cer

1.52

berus

1.13

cers

0.94

ozo

0.93

cery

0.92

pter

0.87

ipher

0.85

atonin

0.85

oute

0.83

idable

0.82

Activations Density 0.006%

words related to chemical substances, particularly glycerin

words related to bodily fluids and medical conditions

No Known Activations