INDEX

Explanations

phrases that provide clarification or rephrasing

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize alternative explanations or paraphrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

IDES

-0.69

vas

-0.63

adium

-0.62

ousel

-0.60

onite

-0.59

ted

-0.59

Accessory

-0.59

 Shining

-0.57

 Archdemon

-0.56

 sund

-0.56

POSITIVE LOGITS

mith

0.97

paces

0.82

terday

0.77

ames

0.71

pace

0.68

 Esports

0.65

poons

0.65

bek

0.65

esa

0.64

esides

0.64

Activations Density 0.012%

phrases that provide clarification or rephrasing

phrases that emphasize alternative explanations or paraphrases

No Comments

No Known Activations