INDEX

Explanations

references to "name" and its variations in coding or data structure contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_4/width_65k/average_l0_66

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.4.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

 name

-0.75

 Name

-0.65

 fallu

-0.60

makeText

-0.57

NAME

-0.56

 navn

-0.54

name

-0.54

 NAME

-0.53

WithName

-0.53

 Names

-0.52

POSITIVE LOGITS

ValueStyle

1.05

styleType

0.75

OGND

0.71

 المعيارى

0.70

ungeon

0.69

 itſelf

0.68

"]:

0.65

 ſche

0.63

"]

0.63

"],

0.63

Activations Density 0.030%

references to "name" and its variations in coding or data structure contexts

No Comments

No Known Activations