INDEX

Explanations

phrases or words related to size, especially emphasizing things that are larger than usual

oai_token-act-pair · gpt-3.5-turbo

terms related to size classification

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

onso

-0.79

WP

-0.76

ance

-0.74

route

-0.74

ario

-0.73

sky

-0.73

SEA

-0.73

ources

-0.72

vironment

-0.71

RP

-0.71

POSITIVE LOGITS

 sized

1.01

 intestine

0.81

 chunks

0.81

 scale

0.80

 stature

0.80

 circumference

0.72

 chunk

0.71

 sizes

0.70

gest

0.70

 versions

0.69

Activations Density 0.019%

phrases or words related to size, especially emphasizing things that are larger than usual

terms related to size classification

No Comments

No Known Activations