INDEX

Explanations

terms related to improving or enhancing quality or outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

CardHeader

-0.51

 osobow

-0.48

RetentionPolicy

-0.48

 Schatz

-0.47

 koszulka

-0.47

 BorderRadius

-0.47

lexikon

-0.46

ț

-0.46

transQ

-0.46

quirrels

-0.46

POSITIVE LOGITS

improve

0.88

improving

0.87

 Improvement

0.85

Improvement

0.85

 Improve

0.85

Improve

0.84

 Mejora

0.83

Improving

0.82

 Improving

0.82

improved

0.81

Activations Density 0.149%

terms related to improving or enhancing quality or outcomes

No Comments

No Known Activations