INDEX

Explanations

proper nouns related to hotels or lodgings

oai_token-act-pair · gpt-3.5-turbo

references to various inns

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

uyomi

-0.82

=-=-=-=-=-=-=-=-

-0.78

 litter

-0.68

vich

-0.65

 Marketable

-0.65

¥

-0.65

 follow

-0.63

_-

-0.62

 Stafford

-0.62

 force

-0.61

POSITIVE LOGITS

Inn

1.35

keeper

1.04

umerable

1.03

ocent

0.94

keepers

0.90

itial

0.85

ovo

0.83

ipedia

0.82

acle

0.81

olen

0.81

Activations Density 0.006%

proper nouns related to hotels or lodgings

references to various inns

No Comments

No Known Activations