INDEX

Explanations

code-related terms and structures related to defining rules and classes

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

 correctes

-0.68

 ***!

-0.63

ViewImports

-0.61

ServletConfig

-0.61

帖最后由

-0.61

expandindo

-0.59

CppMethod

-0.57

-------------</

-0.55

WriteBarrier

-0.54

出版年

-0.54

POSITIVE LOGITS

(::

0.65

("${

0.63

?:

0.56

!!.

0.54

kotlin

0.52

"${

0.52

 Ottawa

0.51

shr

0.51

 kotlin

0.51

<*

0.51

Activations Density 0.165%

code-related terms and structures related to defining rules and classes

No Comments

No Known Activations