INDEX

Explanations

numbers and associated data

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AsUp

-0.77

 Numerade

-0.65

 pinulongan

-0.65

richTextPanel

-0.65

DockStyle

-0.64

Rptr

-0.63

BufferException

-0.63

 समीक्षाओं

-0.61

UnusedPrivate

-0.61

mtrl

-0.61

POSITIVE LOGITS

+:+

0.41

一样

0.41

 hacerlo

0.40

 similarly

0.38

不行

0.38

一樣

0.35

 ones

0.34

 ebenfalls

0.31

 podob

0.31

 Similarly

0.31

Activations Density 0.044%

numbers and associated data

similarly

No Comments

No Known Activations