INDEX

Explanations

names of people or places related to various countries

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names and places

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

ruary

-0.86

 sidx

-0.67

kefeller

-0.66

inki

-0.65

ogle

-0.65

espie

-0.60

CTR

-0.59

 shove

-0.58

eret

-0.58

ournals

-0.58

POSITIVE LOGITS

oshenko

0.81

vic

0.71

Ħ¢

0.68

ova

0.64

 guiActiveUnfocused

0.61

kaya

0.61

 Graphics

0.61

 Prix

0.61

 Duchess

0.59

tiny

0.59

Activations Density 1.272%

names of people or places related to various countries

proper nouns, particularly names and places

No Comments

No Known Activations