INDEX

Explanations

locations or proper nouns

oai_token-act-pair · gpt-3.5-turbo

proper nouns and references to specific organizations, places, or groups

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

Reloaded

-0.63

igious

-0.60

iblical

-0.57

ModLoader

-0.56

risome

-0.56

oidal

-0.55

 mainland

-0.53

announced

-0.53

asonic

-0.53

 eyed

-0.52

POSITIVE LOGITS

wings

0.70

eworks

0.67

][

0.64

.–

0.60

Hes

0.59

quez

0.59

 Intermediate

0.58

sts

0.57

kh

0.57

 Metallic

0.57

Activations Density 0.821%

locations or proper nouns

proper nouns and references to specific organizations, places, or groups

No Comments

No Known Activations