INDEX

Explanations

instances of learning or teaching concepts and experiences

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

 surla

-0.87

ifrance

-0.77

ConstraintMaker

-0.76

Filmografia

-0.75

tvguidetime

-0.75

ThemeOverlay

-0.72

oliver

-0.72

 ddelweddau

-0.72

 MotionEvent

-0.70

 resourceCulture

-0.70

POSITIVE LOGITS

 learn

1.94

 learned

1.88

 LEARN

1.84

 learns

1.82

 Learn

1.80

Learn

1.76

learn

1.66

learned

1.63

LEARN

1.58

 Learned

1.52

Activations Density 0.074%

instances of learning or teaching concepts and experiences

No Comments

No Known Activations