INDEX

Explanations

events or actions happening in a specific context

oai_token-act-pair · gpt-3.5-turbo

significant events or occurrences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

pse

-0.77

oco

-0.65

Reply

-0.64

BuyableInstoreAndOnline

-0.62

ELY

-0.60

 Genie

-0.59

 Genius

-0.57

NG

-0.57

ãĥ³ãĤ¸

-0.56

 Zeal

-0.56

POSITIVE LOGITS

 expires

0.96

 airs

0.87

 began

0.82

 aired

0.78

 arrives

0.78

 ended

0.77

 concludes

0.75

 erupted

0.75

 collapses

0.75

 fades

0.74

Activations Density 0.284%

events or actions happening in a specific context

significant events or occurrences

No Comments

No Known Activations