INDEX

Explanations

phrases related to watching or viewing video content

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Watch" in various contexts

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

bably

-0.76

emia

-0.73

cffffcc

-0.69

ylum

-0.69

 comprom

-0.68

cffff

-0.67

ctrl

-0.64

 inval

-0.63

 headaches

-0.63

 suscept

-0.63

POSITIVE LOGITS

tower

1.22

dog

1.19

dogs

1.08

 clips

0.95

ing

0.94

 Below

0.91

 Dogs

0.90

 footage

0.89

ESPN

0.89

 Video

0.88

Activations Density 0.029%

phrases related to watching or viewing video content

instances of the word "Watch" in various contexts

No Comments

No Known Activations