INDEX

Explanations

references to configuration settings and related adjustments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 force

-0.41

 void

-0.41

red

-0.40

jados

-0.39

 dark

-0.38

 train

-0.38

 courtesy

-0.38

fit

-0.38

 final

-0.37

lay

-0.36

POSITIVE LOGITS

Settings

0.76

IndentedString

0.72

settings

0.72

 Settings

0.65

fromnode

0.62

----</

0.60

dafx

0.59

Exercise

0.59

 كومونز

0.59

 settings

0.59

Activations Density 0.534%

references to configuration settings and related adjustments

No Comments

No Known Activations