INDEX

Explanations

product ratings indicated as 'out of 5' and mentioning specific numbers such as '4' or '5'

oai_token-act-pair · gpt-3.5-turbo

evaluation ratings and their context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

RAW

-0.83

milo

-0.80

Advertisements

-0.78

ername

-0.78

staking

-0.77

yrights

-0.74

phabet

-0.74

 congr

-0.71

osate

-0.68

yi

-0.68

POSITIVE LOGITS

 Boone

0.61

 Buffalo

0.60

blu

0.59

CTR

0.58

 Milwaukee

0.57

Buyable

0.56

 turkey

0.56

 arri

0.56

Tik

0.55

 Tart

0.54

Activations Density 0.122%

product ratings indicated as 'out of 5' and mentioning specific numbers such as '4' or '5'

evaluation ratings and their context

No Comments

No Known Activations