INDEX

Explanations

phrases indicating causation or driving force behind an action or outcome

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate causation or underlying motivations

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

iannopoulos

-0.77

çĦ

-0.76

ereo

-0.75

aido

-0.72

iao

-0.72

roma

-0.71

eon

-0.71

oho

-0.68

 Murdoch

-0.68

achu

-0.67

POSITIVE LOGITS

 driven

0.95

 driving

0.89

driving

0.83

vable

0.71

driven

0.71

Dri

0.70

 passionately

0.70

riers

0.69

ousel

0.68

 drove

0.67

Activations Density 0.020%

phrases indicating causation or driving force behind an action or outcome

phrases that indicate causation or underlying motivations

No Comments

No Known Activations