INDEX

Explanations

phrases related to the return values of functions or methods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

returns

-1.79

 returns

-1.77

 Returns

-1.76

Returns

-1.72

 RETURNS

-1.45

RETURNS

-1.02

 returned

-0.99

returned

-0.91

Returning

-0.90

 returning

-0.89

POSITIVE LOGITS

atimes

0.56

ArgumentParser

0.51

ôn

0.47

 physique

0.47

 viewDidLoad

0.47

LETE

0.47

lück

0.46

delo

0.45

jod

0.45

HideFlags

0.44

Activations Density 0.224%

phrases related to the return values of functions or methods

No Comments

No Known Activations