INDEX

Explanations

references to a person or place with the name "Cow."

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Cow" 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 @ 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

atically

-0.91

ģĸ

-0.73

blance

-0.71

SPA

-0.67

 diplomacy

-0.65

 extrad

-0.65

 planes

-0.65

AMI

-0.64

 Surge

-0.63

sein

-0.62

POSITIVE LOGITS

Cow

3.88

Cow

2.65

cow

2.08

cow

1.82

 Cowboy

1.47

 cows

1.38

 Sheep

1.26

Stew

1.15

 Dairy

1.11

 Goat

1.10

Activations Density 0.019%

references to a person or place with the name "Cow."

mentions of the word "Cow" in various contexts

No Comments

No Known Activations