INDEX

Explanations

mentions of UC (University of California) followed by a numerical value (such as UC 9)

oai_token-act-pair · gpt-3.5-turbo

references to the University of California (UC) system

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

*/(

-0.73

igree

-0.71

 separating

-0.63

driving

-0.62

powers

-0.62

ems

-0.61

 whats

-0.61

alty

-0.61

giving

-0.60

hab

-0.60

POSITIVE LOGITS

UC

1.07

UC

0.96

 Irvine

0.91

IPM

0.89

MJ

0.86

 campuses

0.80

uates

0.78

 Berkeley

0.78

ANN

0.78

UCS

0.75

Activations Density 0.006%

mentions of UC (University of California) followed by a numerical value (such as UC 9)

references to the University of California (UC) system

No Comments

No Known Activations