INDEX

Explanations

phrases indicating a specific or unique individual or item

oai_token-act-pair · gpt-3.5-turbo

repeated phrases, specifically the word "One."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hips

-0.67

="#

-0.65

respective

-0.63

ivas

-0.62

nas

-0.62

osponsors

-0.62

aeda

-0.60

 Vaugh

-0.59

agents

-0.59

 declarations

-0.59

POSITIVE LOGITS

 Hundred

1.15

 hundred

1.12

 Piece

1.05

Shot

0.99

esan

0.92

 thousand

0.91

 Thousand

0.90

 Month

0.88

Drive

0.86

 month

0.85

Activations Density 0.083%

phrases indicating a specific or unique individual or item

repeated phrases, specifically the word "One."

No Comments

No Known Activations