INDEX

Explanations

constructs related to programming concepts and instructions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Personensuche

-0.93

 defaultstate

-0.73

rawDesc

-0.72

 فريبيس

-0.71

áctenos

-0.71

-0.69

WriteBarrier

-0.69

DockStyle

-0.67

τως

-0.66

TagHelper

-0.66

POSITIVE LOGITS

the

0.59

 những

0.55

una

0.50

 את

0.46

 την

0.46

 divers

0.45

0.44

'}),

0.43

.")

0.42

un

0.42

Activations Density 0.058%

constructs related to programming concepts and instructions

No Comments

No Known Activations