INDEX

Explanations

references to masks

oai_token-act-pair · gpt-3.5-turbo

references to masks and their various representations or contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

GGGGGGGG

-0.76

lished

-0.74

ALLY

-0.71

ally

-0.68

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.68

 FANT

-0.67

athan

-0.67

rian

-0.66

APH

-0.66

scill

-0.66

POSITIVE LOGITS

 masks

1.16

 mask

1.15

resses

0.99

 Mask

0.98

mask

0.89

olin

0.85

 concealed

0.78

Mask

0.76

 wearer

0.76

lain

0.75

Activations Density 0.010%

references to masks

references to masks and their various representations or contexts

No Comments

No Known Activations