INDEX

Explanations

mentions of op-eds or opinion pieces

oai_token-act-pair · gpt-3.5-turbo

references to organizations and their affiliations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

thirds

-0.62

 fines

-0.57

POW

-0.57

 Runes

-0.56

 liabilities

-0.56

 breathing

-0.56

 subjects

-0.55

 unlocks

-0.55

AMS

-0.55

 confidentiality

-0.55

POSITIVE LOGITS

lisher

0.89

Thumbnail

0.83

tery

0.80

leck

0.80

monary

0.78

estate

0.77

entimes

0.77

eworthy

0.73

ciation

0.72

Cola

0.72

Activations Density 0.062%

mentions of op-eds or opinion pieces

references to organizations and their affiliations

No Comments

No Known Activations