INDEX

Explanations

the word "to" followed by a verb

oai_token-act-pair · gpt-3.5-turbo

phrases indicating readiness or willingness to take action

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

_.

-0.91

UA

-0.74

quickShipAvailable

-0.73

mask

-0.70

 Compass

-0.70

hai

-0.70

anon

-0.70

hops

-0.69

hes

-0.67

afety

-0.67

POSITIVE LOGITS

 fulfill

1.00

 accept

0.94

 hear

0.94

 accommodate

0.90

 pursue

0.89

 receive

0.89

 abandon

0.89

 comply

0.89

 embrace

0.89

 give

0.89

Activations Density 0.144%

the word "to" followed by a verb

phrases indicating readiness or willingness to take action

No Comments

No Known Activations