INDEX

Explanations

technical parameters and data attributes in a structured format

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

 purpoſe

-0.98

 Signalez

-0.95

 متعلقه

-0.90

ſelf

-0.90

migrationBuilder

-0.88

 reaſon

-0.88

principalColumn

-0.87

KURZBESCHREIBUNG

-0.86

Efq

-0.86

 myſelf

-0.85

POSITIVE LOGITS

0.56

0.54

he

0.53

0.52

0.51

sal

0.49

den

0.48

Activations Density 0.452%

technical parameters and data attributes in a structured format

No Comments

No Known Activations