INDEX

Explanations

phrases related to significant historical events or moments

oai_token-act-pair · gpt-3.5-turbo

references to significant events in history

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

 Cosponsors

-0.87

pez

-0.67

ById

-0.65

aze

-0.64

yne

-0.63

DERR

-0.62

raid

-0.62

Pool

-0.61

arse

-0.60

racuse

-0.59

POSITIVE LOGITS

 history

1.61

history

1.38

 History

1.29

 humankind

1.25

 mankind

1.22

 civilisation

1.12

 modern

1.07

 human

1.07

History

1.06

 civilization

1.04

Activations Density 0.201%

phrases related to significant historical events or moments

references to significant events in history

No Comments

No Known Activations