INDEX

Explanations

the word "otta."

oai_token-act-pair · gpt-3.5-turbo

the word "otta" and its variations

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

 Genie

-0.62

 Spirit

-0.59

 nost

-0.58

 Camel

-0.58

ahime

-0.57

 provocation

-0.57

 behaviour

-0.56

 tick

-0.56

 Monarch

-0.55

 fore

-0.55

POSITIVE LOGITS

otta

4.39

onna

1.29

otti

1.28

arella

1.19

 gotta

1.17

ota

1.04

itto

1.03

itte

1.02

enez

0.99

arde

0.99

Activations Density 0.016%

the word "otta."

the word "otta" and its variations

No Comments

No Known Activations