INDEX

Explanations

acronyms related to technology and data communication protocols

oai_token-act-pair · gpt-3.5-turbo

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hold

-0.97

taboola

-0.85

âĸ¬

-0.85

Interstitial

-0.80

âĸ¬âĸ¬

-0.78

nces

-0.76

gra

-0.73

 Liberties

-0.71

 Sheen

-0.70

 Antarctic

-0.70

POSITIVE LOGITS

TP

1.09

ython

1.08

ACK

1.02

WD

1.00

ocalypse

1.00

etition

0.98

rint

0.98

ractor

0.95

olitics

0.93

ainted

0.92

Activations Density 0.007%

acronyms related to technology and data communication protocols

No Comments

No Known Activations