INDEX

Explanations

mentions of the word "Mic" followed by a number

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Mic" in various contexts

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

-0.80

ACTED

-0.76

ãģĮ

-0.74

ãĤĤ

-0.73

ãģĤ

-0.71

DragonMagazine

-0.71

BuyableInstoreAndOnline

-0.70

ãģį

-0.69

ãģĦ

-0.68

é»Ĵ

-0.67

POSITIVE LOGITS

rones

1.03

ropolitan

1.02

onductor

1.02

olon

0.95

Mic

0.95

onduct

0.89

rodu

0.89

HAEL

0.88

plates

0.85

roman

0.85

Activations Density 0.007%

mentions of the word "Mic" followed by a number

mentions of the word "Mic" in various contexts

No Comments

No Known Activations