INDEX

Explanations

personal pronouns referring to oneself

oai_token-act-pair · gpt-3.5-turbo

the word "I'm" or its variations throughout the text

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 SetTextColor

-0.72

 Stability

-0.64

 assail

-0.62

 implements

-0.60

 override

-0.60

 deform

-0.59

 Carbuncle

-0.58

 meets

-0.57

inav

-0.57

 exploits

-0.56

POSITIVE LOGITS

agine

1.23

 glad

1.07

 gonna

1.05

 guessing

1.03

 thankful

1.00

 sorry

1.00

 proud

0.99

 grateful

0.96

sorry

0.96

 ashamed

0.96

Activations Density 0.039%

personal pronouns referring to oneself

the word "I'm" or its variations throughout the text

No Comments

No Known Activations