INDEX

Explanations

information related to weight and dimensions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 Савезне

-0.59

∮

-0.51

IContainer

-0.44

pantalones

-0.41

bė

-0.41

harusnya

-0.40

 intima

-0.40

ℤ

-0.40

 Hallen

-0.39

$__

-0.38

POSITIVE LOGITS

 weight

2.90

 weigh

2.58

weight

2.57

Weight

2.45

 Weight

2.44

 weights

2.43

 WEIGHT

2.42

 weighing

2.41

 weighs

2.39

weigh

2.31

Activations Density 0.337%

information related to weight and dimensions

No Comments

No Known Activations