INDEX

Explanations

information about different subjects or topics

oai_token-act-pair · gpt-3.5-turbo

references to information regarding various subjects or entities

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

 teasp

-0.84

KK

-0.84

rang

-0.75

igham

-0.74

 Sabha

-0.69

à¤¾

-0.68

ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ

-0.66

kus

-0.66

alez

-0.65

;;;;;;;;

-0.65

POSITIVE LOGITS

how

0.86

tnc

0.74

 whats

0.70

izoph

0.67

 them

0.65

 what

0.64

 whether

0.63

 topics

0.63

 glaciers

0.62

 criminality

0.61

Activations Density 0.096%

information about different subjects or topics

references to information regarding various subjects or entities

No Comments

No Known Activations