INDEX

Explanations

negative sentiments or uncertainties

oai_token-act-pair · gpt-3.5-turbo

negations and expressions of uncertainty

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

shine

-0.70

DAY

-0.67

rifice

-0.64

onement

-0.64

 landsl

-0.64

 malf

-0.64

 termination

-0.63

ria

-0.63

 miscar

-0.60

adish

-0.60

POSITIVE LOGITS

 acquainted

1.63

 accustomed

1.57

 familiar

1.50

 vers

1.42

 aware

1.42

 fascinated

1.38

 knowledgeable

1.27

 proficient

1.25

 adept

1.22

 fluent

1.21

Activations Density 0.281%

negative sentiments or uncertainties

negations and expressions of uncertainty

No Comments

No Known Activations