INDEX

Explanations

names of a specific person: Howard

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yrim

-0.72

profits

-0.71

fal

-0.69

ktop

-0.69

erala

-0.68

displayText

-0.68

asus

-0.68

terday

-0.66

rx

-0.64

ebin

-0.64

POSITIVE LOGITS

 Stern

1.15

 Schultz

0.90

 Sachs

0.89

 Dean

0.89

 Hughes

0.89

shire

0.87

 Schn

0.81

 Howard

0.81

 Stark

0.80

 Dent

0.80

Activations Density 0.038%

names of a specific person: Howard

No Comments

No Known Activations