INDEX

Explanations

dialogue or speech markers in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 auffi

-0.89

kháu

-0.81

 AppColors

-0.79

berdayakan

-0.77

новниш

-0.75

})();

-0.73

 ་་

-0.70

يكب

-0.69

 purpoſe

-0.69

 fides

-0.69

POSITIVE LOGITS

 lowly

0.60

 kise

0.58

Mhm

0.54

0.53

 makna

0.52

 slightly

0.52

 slowly

0.49

-...

0.48

 smirked

0.47

 Whilst

0.47

Activations Density 0.208%

dialogue or speech markers in the text

No Comments

No Known Activations