INDEX

Explanations

phrases that indicate information was mentioned or explained previously or above

oai_token-act-pair · gpt-3.5-turbo

instances of phrases indicating prior references or mentions

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

rine

-0.63

 Entered

-0.61

ROR

-0.60

 lasted

-0.59

 replica

-0.58

cliffe

-0.57

 Females

-0.57

edia

-0.57

Maker

-0.56

BuyableInstoreAndOnline

-0.56

POSITIVE LOGITS

above

0.79

Quotes

0.71

 proverb

0.67

rolet

0.67

EngineDebug

0.66

[|

0.65

Dir

0.64

 Azerb

0.62

à¼

0.62

ä

0.61

Activations Density 0.180%

phrases that indicate information was mentioned or explained previously or above

instances of phrases indicating prior references or mentions

No Comments

No Known Activations