INDEX

Explanations

warnings or cautions issued by various entities or individuals

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Interstitial

-0.88

anuts

-0.71

animate

-0.71

 ILCS

-0.70

portion

-0.65

nan

-0.63

dx

-0.63

rafted

-0.62

iop

-0.60

AMY

-0.60

POSITIVE LOGITS

warn

0.97

 omin

0.94

warning

0.92

 Warn

0.89

 warnings

0.81

 against

0.80

 warns

0.78

 warn

0.75

ingly

0.74

 caution

0.73

Activations Density 0.025%

warnings or cautions issued by various entities or individuals

No Comments

No Known Activations