INDEX

Explanations

mentions of personal accounts or life stories, particularly in the form of memoirs

oai_token-act-pair · gpt-3.5-turbo

terms related to memoirs and autobiographies

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aid

-0.72

owler

-0.69

evil

-0.67

itcher

-0.66

 disparate

-0.65

olen

-0.65

 Pearce

-0.63

ulic

-0.61

 kindred

-0.61

 Hannity

-0.61

POSITIVE LOGITS

 memoir

1.28

 autobiography

1.08

 autobi

1.00

stores

0.83

writer

0.82

writers

0.82

ographies

0.82

ogue

0.82

spection

0.81

oir

0.80

Activations Density 0.011%

mentions of personal accounts or life stories, particularly in the form of memoirs

terms related to memoirs and autobiographies

No Comments

No Known Activations