INDEX

Explanations

words related to data plans

oai_token-act-pair · gpt-3.5-turbo

references to data plans

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 IMAGES

-0.77

atures

-0.69

avez

-0.68

worldly

-0.67

ADRA

-0.67

isks

-0.67

athering

-0.66

phies

-0.66

anny

-0.65

INTON

-0.64

POSITIVE LOGITS

 provider

0.81

 providers

0.81

payer

0.81

uary

0.78

 subscribers

0.76

 reimburse

0.74

holder

0.74

 premiums

0.74

emaker

0.72

 allot

0.71

Activations Density 0.034%

words related to data plans

references to data plans

No Comments

No Known Activations