INDEX

Explanations

comment or documentation headers in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_14/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

principalTable

-0.83

fromnode

-0.77

oredCriteria

-0.75

parsedMessage

-0.74

pecabe

-0.68

 anún

-0.66

Tikang

-0.65

setVerticalGroup

-0.63

genodigd

-0.63

Parcelize

-0.62

POSITIVE LOGITS

the

0.61

///

0.58

//

0.53

0.50

The

0.47

0.45

↑

0.44

The

0.43

/**

0.43

an

0.43

Activations Density 0.020%

comment or documentation headers in code

No Comments

No Known Activations