INDEX

Explanations

mentions of the phrase "Top 9"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "top" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ary

-0.64

naire

-0.62

 Lauder

-0.61

 Baldwin

-0.60

ATIONS

-0.59

Cla

-0.58

yrinth

-0.57

ajor

-0.57

 warr

-0.56

 Gaul

-0.56

POSITIVE LOGITS

ographical

1.25

ography

1.12

ographic

1.08

eka

1.03

deck

1.01

ographically

1.01

most

1.00

ology

0.97

 notch

0.96

ographies

0.94

Activations Density 0.043%

mentions of the phrase "Top 9"

instances of the word "top" in various contexts

No Comments

No Known Activations