INDEX

Explanations

instances where examples are given to clarify a point or make something understandable

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

LAB

-0.73

velt

-0.66

mare

-0.56

heid

-0.56

marine

-0.55

ãĤ©

-0.55

istor

-0.53

ascript

-0.53

lon

-0.53

 BEFORE

-0.52

POSITIVE LOGITS

 instance

1.90

 example

1.89

example

1.23

 starters

1.17

instance

1.07

 Example

0.98

ked

0.98

going

0.94

gery

0.92

geries

0.92

Activations Density 0.074%

instances where examples are given to clarify a point or make something understandable

No Comments

No Known Activations