INDEX

Explanations

terms related to measurement or evaluation 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

 CreateTagHelper

-0.59

<unused3>

-0.57

<unused41>

-0.57

<unused79>

-0.57

<unused8>

-0.57

[@BOS@]

-0.57

گاب

-0.57

<unused43>

-0.57

<unused16>

-0.57

<pad>

-0.57

POSITIVE LOGITS

 Verpflichtung

0.42

مصادر

0.40

About

0.39

د

0.38

 administracy

0.38

 okazji

0.37

The

0.36

En

0.36

 Geschäfts

0.35

Activations Density 0.103%

terms related to measurement or evaluation in various contexts

No Comments

No Known Activations