INDEX

Explanations

phrases emphasizing the significance or encompassing nature of something

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "everything" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

gee

-0.75

idon

-0.73

ename

-0.71

eln

-0.69

eder

-0.64

geons

-0.64

isha

-0.63

odan

-0.63

ker

-0.61

phans

-0.61

POSITIVE LOGITS

 else

1.49

 imaginable

1.32

Else

1.02

 conceivable

1.01

 Else

1.01

THING

0.95

 except

0.92

 toget

0.84

else

0.79

igator

0.78

Activations Density 0.035%

phrases emphasizing the significance or encompassing nature of something

references to the concept of "everything" in various contexts

No Comments

No Known Activations