INDEX

Explanations

phrases related to providing guidance, instructions, or encouragement to others

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

 Firstly

-0.71

 nutshell

-0.69

anny

-0.66

velop

-0.62

entin

-0.61

 Frie

-0.61

Owl

-0.61

 whichever

-0.58

 Guarant

-0.58

 Âł Âł Âł Âł Âł Âł Âł Âł

-0.58

POSITIVE LOGITS

 similarly

1.50

 similar

1.35

 likewise

1.18

 equally

1.13

 same

0.93

similar

0.91

same

0.90

Similar

0.89

 emulate

0.87

 comparable

0.85

Activations Density 0.575%

phrases related to providing guidance, instructions, or encouragement to others

No Comments

No Known Activations