INDEX

Explanations

sections and definitions related to medical or health concepts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RegressionTest

-0.74

⅘

-0.59

PerformLayout

-0.58

+#+

-0.57

 queſta

-0.55

FunctionFlags

-0.54

 snippetHide

-0.53

 BoxFit

-0.52

 canst

-0.52

 ſind

-0.51

POSITIVE LOGITS

 termes

0.45

 term

0.43

 begre

0.43

Definition

0.42

什么是

0.41

 Begriffe

0.41

Definitions

0.39

NameInMap

0.39

 Definitions

0.39

SerializedName

0.38

Activations Density 0.019%

sections and definitions related to medical or health concepts

No Comments

No Known Activations