INDEX

Explanations

references to robots or robot-related activities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "robot."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Beg

-0.78

ippi

-0.72

uries

-0.71

creen

-0.70

laus

-0.69

mary

-0.66

 Birch

-0.66

dn

-0.65

WAYS

-0.65

Neh

-0.63

POSITIVE LOGITS

 robot

0.86

ically

0.84

anical

0.83

 mascot

0.81

 bots

0.77

 swarm

0.74

 Robots

0.74

ichick

0.73

 robots

0.72

zeb

0.71

Activations Density 0.014%

references to robots or robot-related activities

occurrences of the word "robot."

No Comments

No Known Activations