INDEX

Explanations

first-person pronouns ('I') followed by verbs or modal verbs

oai_token-act-pair · gpt-3.5-turbo

occurrences of the pronoun "I" and related expressions of personal opinion or uncertainty

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Georgian

-0.63

heres

-0.61

pires

-0.61

 Gale

-0.60

 centres

-0.58

ictive

-0.57

 Thrones

-0.57

 Boat

-0.56

 optics

-0.56

 Sacrament

-0.56

POSITIVE LOGITS

'm

1.50

am

1.17

xtap

1.13

've

1.01

verson

1.00

 myself

0.99

zzo

0.93

ggy

0.93

aido

0.92

oan

0.92

Activations Density 0.160%

first-person pronouns ('I') followed by verbs or modal verbs

occurrences of the pronoun "I" and related expressions of personal opinion or uncertainty

No Comments

No Known Activations