INDEX

Explanations

verbs indicating actions or behaviors

oai_token-act-pair · gpt-3.5-turbo

instances of the verb "to be" in various forms and tenses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 truths

-0.64

eworld

-0.63

 Nept

-0.62

 rivals

-0.60

arching

-0.59

effic

-0.58

 Columb

-0.58

ston

-0.56

 Modes

-0.56

strings

-0.56

POSITIVE LOGITS

 nominate

0.98

 create

0.96

 educate

0.95

 waive

0.94

 allocate

0.94

 emulate

0.94

 modify

0.93

 prepare

0.93

 revise

0.93

 rename

0.93

Activations Density 0.112%

verbs indicating actions or behaviors

instances of the verb "to be" in various forms and tenses

No Comments

No Known Activations