INDEX

Explanations

instances of the word "be" followed by an adjective or past participle

oai_token-act-pair · gpt-3.5-turbo

variations of the verb "to be" in its different forms

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

Â¶

-0.71

Ong

-0.66

iasco

-0.64

pered

-0.62

 Utilities

-0.61

 Installation

-0.60

 Telegram

-0.60

 Extension

-0.59

Lux

-0.59

 Inventory

-0.59

POSITIVE LOGITS

 tempted

1.33

 able

1.33

 surprised

1.27

 amazed

1.19

 ashamed

1.15

 forgiven

1.14

 inclined

1.11

 astonished

1.06

 shocked

1.03

 thankful

1.03

Activations Density 0.121%

instances of the word "be" followed by an adjective or past participle

variations of the verb "to be" in its different forms

No Comments

No Known Activations