INDEX

Explanations

mentions of specific names

oai_token-act-pair · gpt-3.5-turbo

references to Michelin-rated restaurants and related experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Belt

-0.75

 Tate

-0.71

inventoryQuantity

-0.69

 Kali

-0.69

ERC

-0.67

 Heisman

-0.65

 Hera

-0.64

 Oval

-0.63

ukong

-0.62

urities

-0.62

POSITIVE LOGITS

aton

0.83

phis

0.81

gian

0.76

 unden

0.75

lin

0.74

senal

0.72

rette

0.72

jriwal

0.71

½

0.69

dit

0.69

Activations Density 0.015%

mentions of specific names

references to Michelin-rated restaurants and related experiences

No Comments

No Known Activations