INDEX

Explanations

words related to technology and innovation

oai_token-act-pair · gpt-3.5-turbo

Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.0.hook_resid_post_16384_anthropic

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

HuggingFaceFW/fineweb

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ⓧ

-1.38

-1.35

<bos>

-1.34

/**

-1.25

<?

-1.19

/*

-0.95

<?

-0.91

 shuddered

-0.91

 shivered

-0.89

 endow

-0.82

POSITIVE LOGITS

 maroc

1.00

 catég

0.98

 marte

0.97

 balon

0.96

 ananas

0.95

 habile

0.94

 seksi

0.93

 marea

0.89

 karton

0.89

 silikon

0.89

Activations Density 0.477%

words related to technology and innovation

No Known Activations