INDEX

Explanations

proper nouns related to people and places

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals and their roles within a narrative

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

rawdownloadcloneembedreportprint

-0.73

isine

-0.69

Ö¼

-0.69

incial

-0.67

 Allaah

-0.66

ricanes

-0.66

ockey

-0.64

 eSports

-0.64

 Munich

-0.63

 cups

-0.63

POSITIVE LOGITS

Pag

2.51

 Vance

1.56

pag

1.44

Poe

1.35

Vog

1.32

 Benson

1.15

 Draper

1.13

 Pruitt

1.11

 Pebble

1.09

 snag

1.09

Activations Density 0.059%

proper nouns related to people and places

references to specific individuals and their roles within a narrative

No Comments

No Known Activations