INDEX

Explanations

numbers related to margins of error

oai_token-act-pair · gpt-3.5-turbo

references to statistical margins of error in data reporting

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

JECT

-0.74

mine

-0.72

ibur

-0.66

Uz

-0.65

 reborn

-0.65

Rod

-0.64

Gil

-0.63

yer

-0.62

igor

-0.62

wagen

-0.62

POSITIVE LOGITS

 margin

1.16

 margins

1.08

ageddon

0.77

 heights

0.77

asonic

0.77

olicy

0.76

margin

0.76

 separating

0.74

 unbeliev

0.69

antry

0.68

Activations Density 0.018%

numbers related to margins of error

references to statistical margins of error in data reporting

No Comments

No Known Activations