INDEX

Explanations

references to temporal phrases, specifically those indicating time periods

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

Климат

-0.41

&=&

-0.40

&=&\

-0.39

 AppCompatTheme

-0.38

cein

-0.38

</th>

-0.37

obium

-0.37

appBar

-0.36

←

-0.36

&=&

-0.36

POSITIVE LOGITS

 late

1.95

mid

1.54

late

1.52

 early

1.46

 seventies

1.45

 eighties

1.44

 sixties

1.44

 fifties

1.40

Late

1.38

 thirties

1.35

Activations Density 0.327%

references to temporal phrases, specifically those indicating time periods

No Known Activations