INDEX

Explanations

MDA and ADA-related entities

oai_token-act-pair · gpt-3.5-turbo

instances of numerical values or ratings related to discussions or assessments

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

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

aband

-0.76

 endeav

-0.68

cat

-0.66

 Reincarnated

-0.65

inyl

-0.65

stab

-0.64

burn

-0.60

 footing

-0.60

heit

-0.59

 inactive

-0.56

POSITIVE LOGITS

Newsletter

0.88

Indeed

0.85

Asked

0.84

Refer

0.75

Contribut

0.75

ONSORED

0.75

Shape

0.74

Wisconsin

0.74

Walker

0.74

IRO

0.73

Activations Density 0.181%

MDA and ADA-related entities

instances of numerical values or ratings related to discussions or assessments

No Comments

No Known Activations