INDEX

Explanations

names and titles, specifically those with initials

oai_token-act-pair · gpt-3.5-turbo

names and specific identifiers related to people, organizations, and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

Reviewed

-0.95

NK

-0.64

lde

-0.62

votes

-0.60

NRS

-0.59

CTR

-0.59

 organs

-0.58

 centralized

-0.57

 ACTIONS

-0.55

falls

-0.55

POSITIVE LOGITS

llah

0.79

querque

0.78

urai

0.73

ãĤ¦ãĤ¹

0.72

ENA

0.71

Lago

0.69

eming

0.68

uala

0.68

EMENT

0.66

 Odyssey

0.66

Activations Density 0.519%

names and titles, specifically those with initials

names and specific identifiers related to people, organizations, and titles

No Comments

No Known Activations