INDEX

Explanations

the letter "M" with varying emphasis on its frequency in different contexts

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

-0.68

-0.64

-0.50

-0.46

↵

-0.46

fia

-0.43

大利

-0.42

 monoxide

-0.42

POSITIVE LOGITS

 photolibrary

1.00

migrationBuilder

0.99

 виправивши

0.98

 autorytatywna

0.96

 figliu

0.95

 feroit

0.93

ſelves

0.91

 noDo

0.90

 larmes

0.90

 transfieras

0.89

Activations Density 0.021%

the letter "M" with varying emphasis on its frequency in different contexts

No Comments

No Known Activations