INDEX

Explanations

references to mothers

oai_token-act-pair · gpt-3.5-turbo

mentions of 'mom' or references to maternal figures

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

 Railway

-0.78

 principally

-0.69

 Whilst

-0.67

Â£

-0.66

endiary

-0.65

RAFT

-0.63

 railway

-0.62

lihood

-0.62

Russell

-0.62

 recognised

-0.62

POSITIVE LOGITS

mom

3.63

 moms

2.93

Mom

2.61

dad

2.46

 grandma

2.32

mum

2.28

 mother

2.16

Mom

2.10

mom

2.08

 dads

1.96

Activations Density 0.018%

references to mothers

mentions of 'mom' or references to maternal figures

No Comments

No Known Activations