INDEX

Explanations

terms related to the driving force behind actions or events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "driven."

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

iannopoulos

-0.89

itu

-0.81

umbn

-0.81

achu

-0.78

oho

-0.78

atur

-0.72

assium

-0.71

ivas

-0.70

ANN

-0.70

aido

-0.70

POSITIVE LOGITS

 driven

1.28

 driving

0.95

driving

0.93

Dri

0.85

driven

0.80

 drove

0.77

 drives

0.76

vable

0.76

 fueled

0.75

Dri

0.75

Activations Density 0.011%

terms related to the driving force behind actions or events

occurrences of the word "driven."

No Comments

No Known Activations