INDEX

Explanations

discussions about improvements and recommendations for future research or developments

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

しまう

-0.46

archives

-0.46

ΙΑ

-0.43

yun

-0.41

ANGER

-0.41

pannt

-0.41

áculos

-0.40

 setuptools

-0.40

 Harga

-0.39

asky

-0.39

POSITIVE LOGITS

 improvement

2.96

 improvements

2.76

 Improvement

2.64

improvement

2.63

 improve

2.61

improve

2.60

 Improvements

2.52

 improving

2.51

Improvement

2.50

 improved

2.50

Activations Density 0.575%

discussions about improvements and recommendations for future research or developments

No Comments

No Known Activations