INDEX

Explanations

phrases related to execution or evaluation

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

WER

-0.71

 Parenthood

-0.68

Ern

-0.66

Reviewer

-0.63

bird

-0.62

bones

-0.62

ATL

-0.61

¿½

-0.61

 Fifth

-0.61

 Grail

-0.60

POSITIVE LOGITS

uation

1.91

utions

1.80

uating

1.76

uations

1.73

uated

1.66

uting

1.64

ution

1.63

ulatory

1.62

uates

1.52

utable

1.52

Activations Density 0.101%

phrases related to execution or evaluation

No Comments

No Known Activations