INDEX

Explanations

electronics or tech-related product names, especially smartphones and tablets from Samsung like the Galaxy series

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

Not in Any Lists

No Comments

Negative Logits

<bos>

-1.08

 cluster

-0.73

 continue

-0.66

///**

-0.66

 clusters

-0.65

cluster

-0.64

 decided

-0.62

ുറ

-0.62

 also

-0.62

put

-0.61

POSITIVE LOGITS

 lele

1.60

 meis

1.60

 maneu

1.47

 wien

1.43

 increa

1.41

 guarante

1.40

 ordina

1.38

vns

1.37

 ananas

1.37

 haup

1.35

Activations Density 0.089%

electronics or tech-related product names, especially smartphones and tablets from Samsung like the Galaxy series

No Comments

No Known Activations