INDEX

Explanations

references to different levels of authority or legality, particularly in an international context

oai_token-act-pair · gpt-3.5-turbo

instances of conditional or comparative phrases involving organizations and legal terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Lot

-0.75

Equ

-0.70

UC

-0.66

BUR

-0.66

Asset

-0.63

Vol

-0.63

Ward

-0.62

ACK

-0.60

bub

-0.59

iP

-0.59

POSITIVE LOGITS

ModLoader

0.73

ËĪ

0.67

eanor

0.65

0.64

 cohorts

0.64

rency

0.64

/+

0.63

rogens

0.61

 consequ

0.61

)=(

0.60

Activations Density 0.148%

references to different levels of authority or legality, particularly in an international context

instances of conditional or comparative phrases involving organizations and legal terms

No Comments

No Known Activations