INDEX

Explanations

phrases related to the amount of time a task will require

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the duration or process of completing various actions

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

 david

-0.69

 agre

-0.69

 Smile

-0.68

holm

-0.68

 advertised

-0.64

eers

-0.62

 hates

-0.60

Cong

-0.59

iege

-0.58

vine

-0.58

POSITIVE LOGITS

aways

1.01

 advantage

0.94

 precedence

0.90

 care

0.87

overs

0.85

FINE

0.84

 aback

0.83

ume

0.80

arnaev

0.78

inka

0.78

Activations Density 0.082%

phrases related to the amount of time a task will require

phrases indicating the duration or process of completing various actions

No Comments

No Known Activations