INDEX

Explanations

references to URLs and URIs in a structured format, likely for metadata or resource identification

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

 Infórmanos

-0.61

enumii

-0.42

 HasFactory

-0.39

uitton

-0.39

توض

-0.39

 зо

-0.39

Clik

-0.37

 تعدى

-0.36

StandardCharsets

-0.36

AddTagHelper

-0.36

POSITIVE LOGITS

Хьажоргаш

0.50

Datuak

0.49

 VIAF

0.49

 propOrder

0.47

 lenker

0.47

 AssemblyVersion

0.46

 <=",

0.45

 gynhyrchwyd

0.45

Джерела

0.41

ApiModelProperty

0.39

Activations Density 0.005%

references to URLs and URIs in a structured format, likely for metadata or resource identification

No Comments

No Known Activations