INDEX

Explanations

the word "Ten" repeated in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to the term "Ten" or its variations in multiple contexts

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

****************

-0.71

unct

-0.68

Downloadha

-0.67

cffffcc

-0.67

EStream

-0.66

 mileage

-0.65

lycer

-0.62

ready

-0.62

à¼

-0.62

ulating

-0.61

POSITIVE LOGITS

zin

1.10

acity

1.00

 Thousand

0.96

acious

0.88

aciously

0.85

ney

0.82

essee

0.81

eta

0.80

Ten

0.80

oton

0.78

Activations Density 0.006%

the word "Ten" repeated in various contexts

references to the term "Ten" or its variations in multiple contexts

No Comments

No Known Activations