INDEX

Explanations

locations or organizations with a focus on geopolitical or military aspects

oai_token-act-pair · gpt-3.5-turbo

references to the United States and its involvement in various contexts

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

 showers

-0.71

 revers

-0.71

 headlights

-0.69

 thumbnail

-0.69

 STATS

-0.68

 cram

-0.67

 footnote

-0.67

 Graveyard

-0.66

 balls

-0.66

 conjecture

-0.66

POSITIVE LOGITS

based

1.46

trained

1.40

backed

1.35

Israeli

1.34

imposed

1.34

made

1.33

controlled

1.32

speaking

1.31

style

1.30

centric

1.30

Activations Density 0.037%

locations or organizations with a focus on geopolitical or military aspects

references to the United States and its involvement in various contexts

No Comments

No Known Activations