INDEX

Explanations

phrases related to specific locations or entities, particularly including the word "Culver."

oai_token-act-pair · gpt-3.5-turbo

references to culinary topics or establishments

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Butterfly

-0.82

 Twain

-0.76

*/(

-0.68

ovych

-0.68

 promise

-0.64

 Monsanto

-0.62

 Pathfinder

-0.60

 Gutenberg

-0.60

 Buffett

-0.58

 Bayer

-0.58

POSITIVE LOGITS

Cul

1.05

entric

0.97

inary

0.94

iple

0.93

seys

0.89

mination

0.84

isine

0.80

ãĤ¢ãĥ«

0.80

ueless

0.79

aware

0.78

Activations Density 0.004%

phrases related to specific locations or entities, particularly including the word "Culver."

references to culinary topics or establishments

No Comments

No Known Activations