INDEX

Explanations

terms related to programming structures and operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

stdc

-0.46

-0.39

↵

-0.36

 also

-0.35

-0.34

-0.32

De

-0.32

-0.31

fxml

-0.31

De

-0.30

POSITIVE LOGITS

 betweenstory

0.86

TagMode

0.83

 faſt

0.79

transQ

0.78

expandindo

0.77

 ModelExpression

0.77

 ſtate

0.77

fromnode

0.77

 pleaſure

0.76

 indígen

0.75

Activations Density 0.197%

terms related to programming structures and operations

No Comments

No Known Activations