INDEX

Explanations

phrases indicating contrast or contradiction

oai_token-act-pair · gpt-3.5-turbo

phrases that contain negative statements or contradictions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ebted

-0.64

icka

-0.61

arius

-0.61

ierce

-0.59

 watches

-0.57

ukong

-0.56

miah

-0.55

rupt

-0.55

itialized

-0.54

sense

-0.54

POSITIVE LOGITS

 irrelevant

0.95

©¶æ

0.92

 outwe

0.89

 unlikely

0.86

 hardly

0.83

 beside

0.82

 merely

0.80

livion

0.77

 peanuts

0.76

 moot

0.75

Activations Density 0.131%

phrases indicating contrast or contradiction

phrases that contain negative statements or contradictions

No Comments

No Known Activations