INDEX

Explanations

variations of the word "ion" and related terms

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ion

-2.55

ION

-1.98

ion

-1.75

ions

-1.74

Ion

-1.47

Ion

-1.35

ión

-1.26

IONS

-1.23

iones

-1.18

ione

-1.13

POSITIVE LOGITS

SourceChecksum

0.84

AndEndTag

0.83

RenderAtEndOf

0.80

:✨

0.75

withIdentifier

0.75

 EconPapers

0.74

 onBind

0.74

tanleria

0.71

 ©️

0.71

ImageContext

0.69

Activations Density 0.172%

variations of the word "ion" and related terms

No Comments

No Known Activations