INDEX

Explanations

words related to events happening

oai_token-act-pair · gpt-3.5-turbo

references to events or occurrences within specific timeframes or locations

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

Released

-0.75

beard

-0.72

orah

-0.71

shaw

-0.69

cig

-0.69

haw

-0.68

friends

-0.67

ega

-0.67

 endors

-0.67

 hairc

-0.66

POSITIVE LOGITS

ï¸

0.78

 limbo

0.69

 differently

0.68

 æľ

0.65

 millisec

0.65

 Whitman

0.63

Roe

0.63

 Wonderland

0.63

 Kafka

0.62

 Lange

0.62

Activations Density 0.293%

words related to events happening

references to events or occurrences within specific timeframes or locations

No Comments

No Known Activations