INDEX

Explanations

quantifiable measurements or numerical data points

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 medallion

-0.42

sign

-0.41

там

-0.41

 subunits

-0.41

msub

-0.40

 bedo

-0.40

Portail

-0.40

ue

-0.40

拼音

-0.39

wój

-0.39

POSITIVE LOGITS

 BoxFit

0.90

UnusedPrivate

0.84

 InputDecoration

0.82

 bezeichneter

0.79

ViewFeatures

0.77

msgTypes

0.74

 AssemblyProduct

0.73

 kasarigan

0.72

Datuak

0.71

InitVars

0.70

Activations Density 0.141%

quantifiable measurements or numerical data points

No Comments

No Known Activations