INDEX

Explanations

phrases or words related to flags being raised or issues being highlighted

oai_token-act-pair · gpt-3.5-turbo

instances of the word "flagged" and its related forms, indicating issues or alerts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Strongh

-1.01

elf

-0.78

avers

-0.70

imates

-0.69

 Telesc

-0.67

ists

-0.66

arij

-0.66

asion

-0.65

 rall

-0.63

perty

-0.62

POSITIVE LOGITS

 flagged

0.99

ging

0.88

lights

0.76

ged

0.73

Benz

0.70

leased

0.67

outing

0.66

uned

0.64

idon

0.64

undown

0.64

Activations Density 0.016%

phrases or words related to flags being raised or issues being highlighted

instances of the word "flagged" and its related forms, indicating issues or alerts

No Comments

No Known Activations