INDEX

Explanations

phrases indicating statements or claims made by sources

oai_token-act-pair · gpt-3.5-turbo

phrases indicating assertions or statements made by others

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Balanced

-0.61

enegger

-0.60

ãĤ©

-0.58

EStreamFrame

-0.57

ettlement

-0.56

ugu

-0.55

TG

-0.54

ierra

-0.53

izo

-0.53

 Competitive

-0.53

POSITIVE LOGITS

 goodbye

0.97

 aloud

0.80

 anecd

0.76

 sarcast

0.74

 farewell

0.72

to

0.71

 loudly

0.67

 hello

0.66

dylib

0.65

doms

0.64

Activations Density 0.058%

phrases indicating statements or claims made by sources

phrases indicating assertions or statements made by others

No Comments

No Known Activations