INDEX

Explanations

terms related to numerical ranges or options

oai_token-act-pair · gpt-3.5-turbo

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

 Canaver

-0.61

 Lovecraft

-0.60

ians

-0.58

 Kung

-0.57

 pandemonium

-0.57

 satisfaction

-0.55

 Rodrigo

-0.55

 Lanka

-0.55

 Beauty

-0.55

 hurd

-0.54

POSITIVE LOGITS

thirds

1.73

dimensional

1.24

hander

1.21

sided

1.20

 thirds

1.18

handed

1.17

dozen

1.13

faced

1.12

tone

1.10

tier

1.08

Activations Density 0.049%

terms related to numerical ranges or options

No Comments

No Known Activations