INDEX

Explanations

instances of the word "Thomson" occurring in the text

oai_token-act-pair · gpt-3.5-turbo

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Raider

-0.73

 Labrador

-0.71

juven

-0.67

aeper

-0.66

eer

-0.65

liest

-0.63

OOL

-0.63

ggle

-0.63

 Seahawks

-0.62

naire

-0.62

POSITIVE LOGITS

 Thomson

1.49

Thom

0.98

ickson

0.87

 Browne

0.86

 Reuters

0.85

psons

0.85

ciating

0.84

sonian

0.84

istar

0.84

iola

0.80

Activations Density 0.007%

instances of the word "Thomson" occurring in the text

No Comments

No Known Activations