INDEX

Explanations

numerical statistics or data related terms

oai_token-act-pair · gpt-3.5-turbo

instances of the word "posted" or other verbs indicating performance or achievement statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wrong

-0.64

 contrace

-0.63

inventory

-0.62

iw

-0.61

usb

-0.59

BU

-0.58

 lawy

-0.57

oz

-0.57

 Zionism

-0.57

omb

-0.57

POSITIVE LOGITS

 considerable

0.81

idable

0.79

illac

0.78

 favorable

0.78

rocal

0.74

interstitial

0.73

 impressive

0.73

 higher

0.73

 favourable

0.72

 rave

0.71

Activations Density 0.172%

numerical statistics or data related terms

instances of the word "posted" or other verbs indicating performance or achievement statistics

No Comments

No Known Activations