INDEX

Explanations

phrases indicating completion or finalization

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-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.hook_resid_post

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

PathVariable

-0.74

theless

-0.72

déric

-0.70

 McAllister

-0.69

stateProvider

-0.68

 Seitz

-0.67

 jLabel

-0.66

 depoz

-0.65

itext

-0.65

Kai

-0.63

POSITIVE LOGITS

 done

2.50

done

2.17

 Done

2.16

 DONE

2.12

Done

2.10

DONE

1.96

isDone

1.41

 gedaan

1.24

 Donegal

1.17

 undone

1.06

Activations Density 0.102%

phrases indicating completion or finalization

No Comments

No Known Activations