INDEX

Explanations

references to locations and directions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_4/width_65k/average_l0_66

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.4.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

Зноскі

-0.67

ագրություններ

-0.66

 متعلقه

-0.66

 invokingState

-0.61

 సౌకర్య

-0.59

GEBURTSDATUM

-0.58

 AssemblyCulture

-0.57

SPATH

-0.57

ChildScrollView

-0.55

sizeCache

-0.54

POSITIVE LOGITS

 purpoſe

0.63

();)

0.59

tabler

0.58

 somewhere

0.58

 trying

0.55

rxjs

0.55

]$}

0.54

{},

0.54

 Tatsache

0.54

 meant

0.54

Activations Density 0.143%

references to locations and directions

No Comments

No Known Activations