INDEX

Explanations

mentions of specific events or experiences

oai_token-act-pair · gpt-3.5-turbo

phrases indicating previous experiences or lack thereof

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

akia

-0.64

 eventually

-0.61

çīĪ

-0.58

 shortly

-0.57

 staggered

-0.57

 momentarily

-0.56

 abound

-0.54

Mub

-0.54

 occasionally

-0.53

 aided

-0.53

POSITIVE LOGITS

nor

1.20

 anything

0.96

 anywhere

0.96

 anybody

0.96

anything

0.87

yet

0.86

 since

0.82

 anymore

0.79

 anyone

0.79

 EVER

0.78

Activations Density 0.315%

mentions of specific events or experiences

phrases indicating previous experiences or lack thereof

No Comments

No Known Activations