INDEX

Explanations

references to ownership

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "own" in various contexts

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

antle

-0.78

 membr

-0.72

eday

-0.70

dayName

-0.69

 pilgrims

-0.68

orescent

-0.68

urai

-0.67

atform

-0.66

etrical

-0.66

uters

-0.66

POSITIVE LOGITS

Property

0.76

Own

0.74

holding

0.70

bys

0.69

itiz

0.67

 Hearts

0.67

OWN

0.66

 Bans

0.65

 affairs

0.64

 Dominion

0.63

Activations Density 0.018%

references to ownership

occurrences of the word "own" in various contexts

No Comments

No Known Activations