INDEX

Explanations

mentions or observations of specific details in text

oai_token-act-pair · gpt-3.5-turbo

instances of the word "note" and its interactions with various subjects in the document

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

•test

No Comments

Negative Logits

UME

-0.73

 certific

-0.72

agn

-0.67

atown

-0.66

advertising

-0.66

ribe

-0.65

ribes

-0.64

icted

-0.64

asio

-0.63

 adolesc

-0.63

POSITIVE LOGITS

 similarities

1.09

 similarity

0.96

 discrepancies

0.93

 signs

0.88

 inconsistencies

0.87

 resemblance

0.79

 inconsistency

0.75

 differences

0.74

 deviations

0.72

 anomalies

0.71

Activations Density 0.221%

mentions or observations of specific details in text

instances of the word "note" and its interactions with various subjects in the document

No Comments

No Known Activations