INDEX

Explanations

phrases that express a specific type or category of something

oai_token-act-pair · gpt-3.5-turbo

classifications of types or categories of things

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

INA

-0.86

å§«

-0.82

åħī

-0.73

æ©

-0.69

milo

-0.69

æĸ

-0.68

orsi

-0.67

yi

-0.64

ãĥ´

-0.64

boxing

-0.64

POSITIVE LOGITS

liest

0.86

 thing

0.73

natureconservancy

0.66

rouse

0.65

 MacArthur

0.65

 straw

0.65

 understatement

0.64

aminer

0.64

 formula

0.61

 profitable

0.61

Activations Density 0.083%

phrases that express a specific type or category of something

classifications of types or categories of things

No Comments

No Known Activations