INDEX

Explanations

text related to business advertising, contact information, and personnel backgrounds

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.6.hook_resid_post_16384_anthropic_fast_lr

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

HuggingFaceFW/fineweb

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 massacres

-0.99

 disinformation

-0.88

 atrocities

-0.88

 bombings

-0.87

 insurgents

-0.86

corruption

-0.85

 insurgency

-0.85

ABORT

-0.83

 fascism

-0.83

 traitors

-0.83

POSITIVE LOGITS

<bos>

16.62

expandindo

2.57

GEBURTSDATUM

2.52

 betweenstory

2.43

Autoritní

2.37

 Administrativna

2.33

تقاوى

2.31

 dispen

2.31

 encomp

2.28

 intersper

2.23

Activations Density 0.956%

text related to business advertising, contact information, and personnel backgrounds

No Comments

No Known Activations