INDEX

Explanations

phrases related to procedures, performance, and standards in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-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

Enllaces

-0.57

 ویکی‌پدی

-0.54

Jeografia

-0.54

 Infórmanos

-0.49

SBATCH

-0.48

 ModelExpression

-0.48

 defStyleAttr

-0.47

kusen

-0.47

المناصب

-0.47

 محفوظة

-0.46

POSITIVE LOGITS

 ExecuteAsync

0.40

 AspNetCore

0.37

omány

0.37

ApiModelProperty

0.36

 hrá

0.35

 заболевания

0.32

 Measurement

0.32

forChild

0.31

 operazioni

0.31

 measurement

0.31

Activations Density 0.267%

phrases related to procedures, performance, and standards in various contexts

No Comments

No Known Activations