INDEX

Explanations

technical and operational terms, particularly in a programming or analytical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

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

tagHelperRunner

-0.89

RenderAtEndOf

-0.85

TagMode

-0.74

 CreateTagHelper

-0.73

 հղումներ

-0.72

دانشنامهٔ

-0.71

 الرياضيه

-0.70

<unused28>

-0.69

<unused14>

-0.69

<unused8>

-0.69

POSITIVE LOGITS

0.31

 Road

0.30

</b>

0.28

pr

0.28

en

0.28

</strong>

0.27

щу

0.27

full

0.27

BL

0.26

ap

0.26

Activations Density 0.066%

technical and operational terms, particularly in a programming or analytical context

No Comments

No Known Activations