INDEX

Explanations

terms related to categories and classifications in media content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

 mariée

-0.85

 démission

-0.81

 grève

-0.77

 argint

-0.76

 مواليد

-0.75

 Romains

-0.71

 fallu

-0.70

 inoxydable

-0.69

 față

-0.69

 cérami

-0.69

POSITIVE LOGITS

 resourceCulture

0.84

'])->

0.68

 hama

0.67

</caption>

0.66

knapp

0.65

 ''){

0.65

pla

0.64

PON

0.64

')){

0.64

 Abbey

0.64

Activations Density 0.012%

terms related to categories and classifications in media content

No Comments

No Known Activations