INDEX

Explanations

specific paths or courses of action mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a pathway or direction toward achieving a goal

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

notations

-0.87

minus

-0.79

context

-0.77

pool

-0.77

leck

-0.76

notation

-0.73

*/(

-0.71

wic

-0.69

men

-0.69

 partName

-0.67

POSITIVE LOGITS

 victory

1.04

 destruction

1.03

 madness

0.99

 redemption

0.92

 vengeance

0.86

 completion

0.85

 defeat

0.84

 Defeat

0.83

 fulfillment

0.82

 adulthood

0.82

Activations Density 0.255%

specific paths or courses of action mentioned in the text

phrases indicating a pathway or direction toward achieving a goal

No Comments

No Known Activations