INDEX

Explanations

scientific theories

oai_token-act-pair · gpt-3.5-turbo

references to theories in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

elight

-0.70

ments

-0.67

ankind

-0.66

semb

-0.64

igi

-0.63

psey

-0.63

events

-0.63

 consecut

-0.63

âĹ¼

-0.62

emetery

-0.61

POSITIVE LOGITS

craft

0.90

sonian

0.88

 theory

0.81

 underpin

0.80

iple

0.79

ually

0.78

ologies

0.78

icist

0.77

uca

0.76

 Theory

0.75

Activations Density 0.012%

scientific theories

references to theories in various contexts

No Comments

No Known Activations