INDEX

Explanations

code-related constructs and elements within a programming context

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

kháu

-0.42

+:+

-0.38

EndInit

-0.36

OptionsMenu

-0.36

BeginInit

-0.35

\{\\

-0.34

DatabaseError

-0.34

GenerationType

-0.32

 abstrato

-0.32

bitField

-0.32

POSITIVE LOGITS

ValueStyle

0.79

RenderAtEndOf

0.61

BagLayout

0.59

 autorytatywna

0.53

 BoxDecoration

0.52

 Plenum

0.51

LayoutPanel

0.48

%)$

0.48

sidemargin

0.47

 Containing

0.46

Activations Density 0.029%

code-related constructs and elements within a programming context

No Comments

No Known Activations