INDEX

Explanations

the word "never" occurring in sentences

oai_token-act-pair · gpt-3.5-turbo

phrases indicating repeated actions or events that did not occur

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 antioxid

-0.79

ahime

-0.74

=-=-=-=-=-=-=-=-

-0.67

 Crusher

-0.67

OUR

-0.64

âĶģ

-0.61

equality

-0.60

Disclaimer

-0.59

ÃįÃį

-0.59

PI

-0.58

POSITIVE LOGITS

theless

1.59

 bothered

1.21

 ceases

1.13

mind

1.04

 knew

1.02

 imagined

1.02

 ceased

1.02

 doubted

1.02

 dreamed

1.01

 forgot

0.98

Activations Density 0.045%

the word "never" occurring in sentences

phrases indicating repeated actions or events that did not occur

No Comments

No Known Activations