INDEX

Explanations

sequences of repetitive characters or ellipses

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

................

-1.05

..............

-0.65

.............

-0.65

........

-0.64

............

-0.63

...........

-0.62

 ........

-0.61

...............

-0.61

 .............

-0.59

..........

-0.59

POSITIVE LOGITS

MigrationBuilder

0.83

 InputDecoration

0.82

 препратки

0.79

 nahilalakip

0.74

AddTagHelper

0.72

 "..\..\..\

0.72

__":

0.71

 LEYENDO

0.71

 ligiloj

0.70

ientôt

0.68

Activations Density 0.008%

sequences of repetitive characters or ellipses

No Comments

No Known Activations