INDEX

Explanations

expressions related to political decisions and their potential consequences

oai_token-act-pair · gpt-3.5-turbo

conditional phrases indicating potential future outcomes or consequences

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

 guesses

-0.76

noticed

-0.67

 sang

-0.67

told

-0.67

 confessed

-0.66

 mastered

-0.65

 Printed

-0.65

 Reporting

-0.64

 narrated

-0.64

 Named

-0.64

POSITIVE LOGITS

 undermine

1.47

 endanger

1.41

 exacerbate

1.38

 worsen

1.38

 impede

1.38

 jeopard

1.34

 hinder

1.34

 weaken

1.34

 violate

1.33

 devast

1.31

Activations Density 0.206%

expressions related to political decisions and their potential consequences

conditional phrases indicating potential future outcomes or consequences

No Comments

No Known Activations