INDEX

Explanations

keywords related to regulations or rules

oai_token-act-pair · gpt-3.5-turbo

concepts related to new rules, standards, technologies, and evidence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

otal

-0.72

sung

-0.69

ealous

-0.68

ammad

-0.66

formance

-0.66

indu

-0.62

akespeare

-0.60

emo

-0.60

odor

-0.60

 Pett

-0.59

POSITIVE LOGITS

 reinforcements

0.70

ãĥķãĤ©

0.69

 arrivals

0.69

 architectures

0.69

stadt

0.66

Geo

0.66

 hire

0.65

 entrants

0.64

çķ

0.63

 hires

0.63

Activations Density 0.258%

keywords related to regulations or rules

concepts related to new rules, standards, technologies, and evidence

No Comments

No Known Activations