INDEX

Explanations

mentions of rumors or speculations

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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Annotations

-0.68

athetic

-0.66

 uneven

-0.63

activated

-0.62

 personalized

-0.62

assisted

-0.61

athy

-0.61

 humans

-0.60

 human

-0.57

 accredited

-0.57

POSITIVE LOGITS

Rum

1.45

Rum

1.22

uz

0.96

imaru

0.86

ors

0.83

inating

0.81

Boss

0.80

illery

0.79

 unden

0.77

ovie

0.75

Activations Density 0.007%

mentions of rumors or speculations

mentions of rumors

No Comments

No Known Activations