INDEX

Explanations

times expressed in time formats with periods

oai_token-act-pair · gpt-3.5-turbo

instances of periods and punctuation marks

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

isation

-0.73

odor

-0.69

isations

-0.68

·

-0.68

hiba

-0.67

izoph

-0.66

isable

-0.64

igion

-0.64

 resemb

-0.64

chie

-0.63

POSITIVE LOGITS

EDT

1.45

ET

1.36

EST

1.34

PDT

1.32

PST

1.30

CST

1.27

PT

1.20

CT

1.07

CET

1.05

 Eastern

0.99

Activations Density 0.030%

times expressed in time formats with periods

instances of periods and punctuation marks

No Comments

No Known Activations