INDEX

Explanations

the presence of the term "ire" 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 @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

flashdata

-0.63

enumii

-0.61

enumi

-0.58

 ILogger

-0.58

 setw

-0.55

ngdoc

-0.52

 Audit

-0.51

Personensuche

-0.50

 ActionResult

-0.50

 resultList

-0.50

POSITIVE LOGITS

 spørsmål

0.59

 telefónica

0.57

 detección

0.55

 læng

0.53

Quick

0.52

 crece

0.52

 recomendación

0.51

 encendido

0.51

 rodillas

0.51

 crecer

0.51

Activations Density 0.401%

the presence of the term "ire" in various contexts

No Comments

No Known Activations