INDEX

Explanations

names or terms related to specific individuals or places

oai_token-act-pair · gpt-3.5-turbo

words and elements related to medication and health

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

nesday

-0.83

ruary

-0.75

abase

-0.69

ortium

-0.65

 livest

-0.63

risome

-0.63

arcity

-0.61

 nomine

-0.59

 lent

-0.57

ndra

-0.57

POSITIVE LOGITS

INAL

0.88

kov

0.86

yna

0.85

gaard

0.75

ä½ľ

0.75

indu

0.73

 Wrap

0.70

intel

0.70

horn

0.68

opa

0.68

Activations Density 0.239%

names or terms related to specific individuals or places

words and elements related to medication and health

No Comments

No Known Activations