INDEX

Explanations

names, likely of a person or place

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word 'ston'

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

 STATS

-0.69

HIP

-0.68

ãģĤ

-0.67

 tactical

-0.67

 mileage

-0.66

AQ

-0.66

Khe

-0.65

 loopholes

-0.62

Pwr

-0.62

ULE

-0.62

POSITIVE LOGITS

ston

1.40

enberg

1.02

ufact

0.93

stead

0.91

igan

0.89

igans

0.85

enburg

0.85

pton

0.84

ishing

0.83

keye

0.83

Activations Density 0.006%

names, likely of a person or place

occurrences of the word 'ston'

No Comments

No Known Activations