INDEX

Explanations

references to game states or player roles

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

Bioaccumulative

-0.59

 Bioaccumulative

-0.53

RenderAtEndOf

-0.52

calientes

-0.51

 Paglinawan

-0.50

ufficio

-0.49

msgTypes

-0.47

StructEnd

-0.47

sidemargin

-0.47

 Exacts

-0.46

POSITIVE LOGITS

Visitor

0.48

 occup

0.44

 visit

0.42

 renovation

0.40

 inspection

0.40

%>

0.39

 occupation

0.39

 renovating

0.39

 renovations

0.39

 Visit

0.38

Activations Density 0.318%

references to game states or player roles

No Comments

No Known Activations