INDEX

Explanations

numerical representations related to achievements or scores

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

OGND

-0.67

 externi

-0.61

 الرياضيه

-0.58

 onAnimation

-0.55

 ujednoznacz

-0.54

 Analyzed

-0.53

LayoutStyle

-0.52

WebVitals

-0.52

webElementGuid

-0.52

IndentedString

-0.51

POSITIVE LOGITS

 sessionId

0.55

 projectId

0.53

pectives

0.53

ardians

0.53

wirtschaft

0.52

oldown

0.52

 jsonArray

0.52

hexyl

0.52

 stringBuilder

0.51

 newVal

0.51

Activations Density 0.023%

numerical representations related to achievements or scores

No Comments

No Known Activations