INDEX

Explanations

references to degrees or measures of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.hook_mlp_out

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

 PRINCIPAL

-0.58

 noqa

-0.58

ungkapkan

-0.52

Principal

-0.50

principal

-0.49

CompleteListener

-0.48

 Principal

-0.48

 Lenti

-0.47

 korte

-0.46

Linki

-0.46

POSITIVE LOGITS

хьтан

0.68

0.65

Handlung

0.62

 Administrativna

0.61

AccessorTable

0.61

soever

0.59

ніципа

0.57

();)

0.57

}');

0.57

STRA

0.56

Activations Density 0.027%

references to degrees or measures of something

No Comments

No Known Activations