INDEX

Explanations

numeric values and measurements in cooking instructions

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

 lifetimes

-0.53

něž

-0.51

anskje

-0.47

 Wilayah

-0.47

tvguidetime

-0.46

 pregunto

-0.46

iconque

-0.46

hawatir

-0.45

telse

-0.44

 Generations

-0.44

POSITIVE LOGITS

 Taktlose

0.41

 CreateTagHelper

0.41

webElementXpaths

0.41

AxisAlignment

0.40

wapV

0.40

 defaultstate

0.38

 ExecuteAsync

0.38

]=>

0.37

 pinch

0.37

 crushed

0.35

Activations Density 0.002%

numeric values and measurements in cooking instructions

No Comments

No Known Activations