INDEX

Explanations

terms related to rumors and gossip

oai_token-act-pair · gpt-3.5-turbo

mentions of rumors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

lihood

-0.85

 humans

-0.67

 learning

-0.64

 Egyptians

-0.61

 Contrast

-0.60

 Parenthood

-0.60

 Wildcats

-0.58

 human

-0.56

 Cancel

-0.56

 BEST

-0.56

POSITIVE LOGITS

pled

1.18

oured

1.16

atis

1.15

inating

1.14

ble

1.01

inatory

0.99

rum

0.96

undo

0.96

ored

0.95

uel

0.93

Activations Density 0.020%

terms related to rumors and gossip

mentions of rumors

No Comments

No Known Activations