INDEX

Explanations

variations or differences in items or categories

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

nothing

-0.70

ISTORY

-0.69

WARD

-0.68

Never

-0.67

OIL

-0.67

Behind

-0.66

Ĭ

-0.64

never

-0.64

emp

-0.64

Alert

-0.64

POSITIVE LOGITS

iating

2.03

 kinds

1.84

iates

1.72

 types

1.65

iations

1.62

ials

1.58

 versions

1.39

iator

1.36

iated

1.34

 aspects

1.29

Activations Density 0.071%

variations or differences in items or categories

No Comments

No Known Activations