INDEX

Explanations

research reports and claims related to new findings or methodologies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.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

 ComVisible

-0.82

 linkovi

-0.81

DllImport

-0.78

WebVitals

-0.77

 Shakspeare

-0.76

ClientSize

-0.75

oredCriteria

-0.73

PreferredItem

-0.72

CloseOperation

-0.71

 beginnetje

-0.71

POSITIVE LOGITS

 möjligt

0.53

 mengin

0.48

 daarom

0.47

 mulig

0.46

 därför

0.45

 bakom

0.45

 möj

0.44

 vindt

0.44

 utilizó

0.43

 kív

0.43

Activations Density 1.002%

research reports and claims related to new findings or methodologies

No Comments

No Known Activations