INDEX

Explanations

technical terms and methods related to numerical simulations and mathematical modeling

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

degree

-0.32

 COUN

-0.30

 degree

-0.30

новниш

-0.26

Pan

-0.26

transforms

-0.26

distribution

-0.25

 transforma

-0.25

Transform

-0.25

align

-0.25

POSITIVE LOGITS

 '\\;'

0.81

 typelib

0.78

 HasFactory

0.76

yntaxException

0.74

MethodManager

0.73

 ſind

0.70

<unused8>

0.69

<unused17>

0.69

<unused14>

0.69

[@BOS@]

0.69

Activations Density 0.069%

technical terms and methods related to numerical simulations and mathematical modeling

No Comments

No Known Activations