INDEX

Explanations

words related to historical events or figures

oai_token-act-pair · gpt-3.5-turbo

references to historical topics or figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

holder

-0.69

FCC

-0.69

 Wonderland

-0.67

 thumbs

-0.65

BOX

-0.64

 learning

-0.64

 loophole

-0.63

OMG

-0.63

 loose

-0.61

MO

-0.61

POSITIVE LOGITS

orically

1.77

orical

1.66

oric

1.58

orians

1.41

orian

1.38

oria

1.27

adr

1.21

oire

1.14

orie

1.14

ori

1.14

Activations Density 0.039%

words related to historical events or figures

references to historical topics or figures

No Comments

No Known Activations