INDEX

Explanations

references to mozzarella cheese

oai_token-act-pair · gpt-3.5-turbo

references to the word "mo" in various contexts, indicating an emphasis on that term

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ãĥ´ãĤ¡

-0.76

 Drawn

-0.74

scl

-0.70

orius

-0.70

é¾įåĸļå£«

-0.70

AMS

-0.69

lihood

-0.67

mania

-0.66

ç¥ŀ

-0.66

nings

-0.65

POSITIVE LOGITS

ose

0.93

utes

0.84

veland

0.81

oths

0.80

ogly

0.80

aned

0.79

mo

0.79

osh

0.78

oser

0.77

ploy

0.75

Activations Density 0.007%

references to mozzarella cheese

references to the word "mo" in various contexts, indicating an emphasis on that term

No Comments

No Known Activations