INDEX

Explanations

the word "nuts" with varying levels of intensity

oai_token-act-pair · gpt-3.5-turbo

mentions of "nuts" or "peanuts."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Dialog

-0.71

 Friendship

-0.69

 Catherine

-0.68

 Collection

-0.68

separ

-0.68

 Courtesy

-0.66

Ink

-0.65

ime

-0.64

Ven

-0.63

 Transfer

-0.63

POSITIVE LOGITS

 nuts

4.15

nuts

2.45

nut

2.23

 peanuts

1.93

 bananas

1.70

 almonds

1.63

 crazy

1.49

nut

1.43

Nut

1.33

 insane

1.29

Activations Density 0.005%

the word "nuts" with varying levels of intensity

mentions of "nuts" or "peanuts."

No Comments

No Known Activations