INDEX

Explanations

phrases related to restricting or limiting something

oai_token-act-pair · gpt-3.5-turbo

phrases that reference limitations or restrictions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

engine

-0.71

grain

-0.71

SPONSORED

-0.70

 largeDownload

-0.70

negie

-0.67

});

-0.65

endra

-0.63

quickShipAvailable

-0.63

started

-0.63

tun

-0.61

POSITIVE LOGITS

othy

0.78

asty

0.77

 preserve

0.75

 afar

0.72

 accommodate

0.71

pha

0.69

 conserve

0.69

 exclude

0.68

 appease

0.66

 subsistence

0.66

Activations Density 0.102%

phrases related to restricting or limiting something

phrases that reference limitations or restrictions

No Comments

No Known Activations