INDEX

Explanations

quantitative statements expressed as percentages or totals

oai_token-act-pair · gpt-3.5-turbo

references to percentages or statistical measures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ful

-0.68

edd

-0.67

æ³

-0.66

lier

-0.64

haps

-0.63

zai

-0.62

arten

-0.61

fy

-0.60

ALLY

-0.60

smith

-0.60

POSITIVE LOGITS

 populace

0.97

 electorate

0.97

 population

0.97

 country

0.94

 nation

0.93

 overall

0.86

 total

0.85

 workforce

0.84

 equation

0.84

 world

0.79

Activations Density 0.117%

quantitative statements expressed as percentages or totals

references to percentages or statistical measures

No Comments

No Known Activations