INDEX

Explanations

relevant content associated with military operations

oai_token-act-pair · gpt-3.5-turbo

instances of repeated phrases or structures in the text

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

Brexit

-0.77

Scotland

-0.72

Netflix

-0.70

abetes

-0.70

Labour

-0.69

ghazi

-0.69

-0.68

Sexual

-0.68

avid

-0.66

Britain

-0.66

POSITIVE LOGITS

 latter

1.20

 discrepancy

0.98

 arrangement

0.85

 aforementioned

0.84

 initial

0.84

 resultant

0.80

 same

0.80

 difference

0.79

 situation

0.79

 timing

0.79

Activations Density 0.733%

relevant content associated with military operations

instances of repeated phrases or structures in the text

No Comments

No Known Activations