INDEX

Explanations

text segments containing diagrams or schematic illustrations

oai_token-act-pair · gpt-3.5-turbo

references to diagrams and schematics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ierra

-0.73

 Advocate

-0.65

NetMessage

-0.64

VO

-0.64

elope

-0.63

 unden

-0.63

nesota

-0.61

ternity

-0.61

 rainy

-0.60

 foothold

-0.60

POSITIVE LOGITS

matically

1.20

matic

1.10

ming

1.08

 diagram

1.05

matical

1.05

 diagrams

1.03

zag

0.92

urally

0.85

mat

0.81

 outline

0.80

Activations Density 0.026%

text segments containing diagrams or schematic illustrations

references to diagrams and schematics

No Comments

No Known Activations