INDEX

Explanations

numerical values indicating money in dollars

oai_token-act-pair · gpt-3.5-turbo

references to the number 125 and other similar numerical values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 trees

-0.67

jong

-0.67

 Jobs

-0.66

 Brig

-0.66

 Bark

-0.65

naires

-0.65

 Mercenary

-0.64

 bios

-0.63

 thumbs

-0.62

ados

-0.62

POSITIVE LOGITS

2.87

2.16

1.89

1.37

1.31

1.25

1.20

1.17

1.16

Activations Density 0.038%

numerical values indicating money in dollars

references to the number 125 and other similar numerical values

No Comments

No Known Activations