INDEX

Explanations

references to the word "Temple"

oai_token-act-pair · gpt-3.5-turbo

references to "Temple."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

leneck

-0.80

enegger

-0.75

iability

-0.75

ÑĮ

-0.71

agher

-0.69

imil

-0.69

nesota

-0.67

alez

-0.67

Leary

-0.67

ives

-0.66

POSITIVE LOGITS

 temple

0.87

 ceremonies

0.83

 Mount

0.81

 Kush

0.79

 Ruins

0.78

plates

0.78

 temples

0.76

 Trials

0.74

 ceremony

0.74

keeper

0.73

Activations Density 0.043%

references to the word "Temple"

references to "Temple."

No Comments

No Known Activations