INDEX

Explanations

phrases mentioning both men and women

oai_token-act-pair · gpt-3.5-turbo

references to gender, particularly focusing on the roles and presence of women in various contexts

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

word2 in "word1 and word2", especially when the words are conceptually related

Explanation Uploaded by User

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

•L9 resid features following 'and'

No Comments

Negative Logits

Joy

-0.71

Thunder

-0.68

MN

-0.68

Dur

-0.68

Kevin

-0.67

Tok

-0.67

afety

-0.66

San

-0.66

GREEN

-0.65

Inv

-0.65

POSITIVE LOGITS

 alike

1.57

 striped

1.00

 respectively

0.99

 combatants

0.85

 halves

0.69

 faiths

0.67

 sexes

0.66

 separated

0.66

 equally

0.66

 separately

0.65

Activations Density 0.131%

phrases mentioning both men and women

references to gender, particularly focusing on the roles and presence of women in various contexts

word2 in "word1 and word2", especially when the words are conceptually related

No Comments

No Known Activations