INDEX

Explanations

terms related to recognition or being recognized

oai_token-act-pair · gpt-3.5-turbo

words related to recognition and brand awareness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Rail

-0.76

 Grind

-0.72

 Hobby

-0.65

Blog

-0.63

bury

-0.61

 Travel

-0.59

 Heaven

-0.59

 Tant

-0.58

 Transportation

-0.57

 anecdotes

-0.57

POSITIVE LOGITS

isable

1.24

 recogn

1.18

recogn

1.14

 recognise

1.06

izable

1.03

isance

1.02

ibly

0.95

ably

0.90

itaire

0.89

 recognizable

0.88

Activations Density 0.009%

terms related to recognition or being recognized

words related to recognition and brand awareness

No Comments

No Known Activations