INDEX

Explanations

instances of a specific name or title related to a business or individual

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

 Half

-1.11

Half

-0.98

 wikipagina

-0.88

 anún

-0.72

 îna

-0.67

 atardecer

-0.67

vádz

-0.65

 tartalmaz

-0.63

 behaviors

-0.63

 Lots

-0.63

POSITIVE LOGITS

two

0.87

one

0.84

 three

0.74

 five

0.72

 four

0.71

six

0.66

 double

0.65

 second

0.65

few

0.65

THISDAY

0.64

Activations Density 0.005%

instances of a specific name or title related to a business or individual

No Comments

No Known Activations