INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

names of people or characters

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

anwhile

-0.71

etheless

-0.63

mble

-0.62

*/(

-0.61

espie

-0.55

sembly

-0.55

umbnails

-0.54

cffffcc

-0.53

)=(

-0.52

 dism

-0.52

POSITIVE LOGITS

 Street

0.73

 Avenue

0.67

 Beach

0.63

 Creek

0.63

 Island

0.62

 Hill

0.61

eland

0.59

 Lake

0.59

ont

0.58

inia

0.57

Activations Density 0.745%

proper nouns

names of people or characters

No Comments

No Known Activations