INDEX

Explanations

mentions of the word "Prime" in the text

oai_token-act-pair · gpt-3.5-turbo

references to the term "Prime."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

anwhile

-0.77

rought

-0.74

ioch

-0.70

bring

-0.70

ologies

-0.69

enegger

-0.68

ersion

-0.68

anners

-0.68

olester

-0.67

sie

-0.66

POSITIVE LOGITS

 Minister

0.99

 Prime

0.92

 Directive

0.91

knit

0.90

Prime

0.82

 Ministers

0.81

Rib

0.81

 Affairs

0.74

 Recep

0.73

 Narendra

0.71

Activations Density 0.010%

mentions of the word "Prime" in the text

references to the term "Prime."

No Comments

No Known Activations