INDEX

Explanations

assertions in test cases for functions and their expected outputs

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

 FetchType

-0.59

BufferException

-0.57

SBATCH

-0.55

 betweenstory

-0.54

зраст

-0.54

 يتيمه

-0.52

 kaarangay

-0.51

localVar

-0.51

fromnode

-0.51

艋

-0.50

POSITIVE LOGITS

Panamoan

0.42

 contextLoads

0.41

 bóng

0.34

 existence

0.33

 nakalista

0.33

ContentLoaded

0.33

 Wiktionnaire

0.32

 bougies

0.32

setBold

0.32

Autoritní

0.32

Activations Density 0.001%

assertions in test cases for functions and their expected outputs

No Comments

No Known Activations