INDEX

Explanations

terms related to sizes and lengths in various contexts

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

 jambe

-0.40

diğini

-0.39

idu

-0.39

TagHelper

-0.39

mischer

-0.38

 incomplète

-0.38

 Reiche

-0.38

 Fewer

-0.38

yors

-0.38

 AssemblyProduct

-0.38

POSITIVE LOGITS

 sized

2.21

 size

2.08

 level

1.89

sized

1.82

size

1.76

level

1.62

 SIZE

1.49

 length

1.48

 Size

1.43

Sized

1.43

Activations Density 0.395%

terms related to sizes and lengths in various contexts

No Comments

No Known Activations