INDEX

Explanations

numerical values followed by units of measurement

oai_token-act-pair · gpt-3.5-turbo

measurements and units of quantity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Ü

-0.77

itialized

-0.75

ĪĴ

-0.68

actionDate

-0.64

soDeliveryDate

-0.64

 personalities

-0.61

obbies

-0.61

 mentors

-0.60

 acron

-0.60

 Shack

-0.59

POSITIVE LOGITS

 squared

0.93

 basis

0.89

Â²

0.89

 ratio

0.88

 average

0.87

imeter

0.85

 increments

0.85

 multiplied

0.81

 whereas

0.76

/$

0.75

Activations Density 0.105%

numerical values followed by units of measurement

measurements and units of quantity

No Comments

No Known Activations