INDEX

Explanations

the word "ass" at varying levels of activation

oai_token-act-pair · gpt-3.5-turbo

the word "ass" 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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Advertisement

-0.67

 Neph

-0.67

 bite

-0.65

 effic

-0.64

 Britann

-0.64

VID

-0.61

 Welsh

-0.60

SIG

-0.59

 Libyan

-0.59

 break

-0.59

POSITIVE LOGITS

essor

1.41

assin

1.37

oci

1.21

essment

1.19

ociation

1.15

istant

1.13

ortment

1.13

andra

1.09

istance

1.08

imilation

1.08

Activations Density 0.012%

the word "ass" at varying levels of activation

the word "ass" in various contexts

No Comments

No Known Activations