INDEX

Explanations

sequences that are likely non-linguistic characters or encoding artifacts, potentially related to a specific type of coding or processing

oai_token-act-pair · gpt-3.5-turbo

special characters or non-standard symbols

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Beir

-0.79

orkshire

-0.72

 Kern

-0.71

 Anat

-0.69

 guiActiveUn

-0.68

 bonded

-0.68

 Stoke

-0.67

xus

-0.65

 Brist

-0.65

clerosis

-0.64

POSITIVE LOGITS

ļ

1.40

¿

1.33

ĳ

1.24

Ī

1.22

¹

1.20

¼

1.19

¶

1.19

¤

1.18

ı

1.17

¬

1.17

Activations Density 0.003%

sequences that are likely non-linguistic characters or encoding artifacts, potentially related to a specific type of coding or processing

special characters or non-standard symbols

No Comments

No Known Activations