INDEX

Explanations

references to a specific female subject, indicating focus on her actions and experiences

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

torch.float32

Hook Point

blocks.12.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

HuggingFaceFW/fineweb

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-1.72

 constate

-0.68

 stiller

-0.57

<?

-0.56

ⓧ

-0.55

 vinil

-0.54

 sedi

-0.54

 vola

-0.53

 fré

-0.53

EndContext

-0.53

POSITIVE LOGITS

 chrysler

1.11

 shes

1.10

she

1.10

 maneu

1.08

 herself

1.07

 impra

1.02

She

0.99

herself

0.98

 Minang

0.97

she

0.96

Activations Density 0.187%

references to a specific female subject, indicating focus on her actions and experiences

No Comments

No Known Activations