INDEX

Explanations

metadata or identifiers associated with document formatting or references

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

afficheront

-0.80

 Infórmanos

-0.65

новниш

-0.64

 ब्रेकडाउन

-0.59

 mostrarán

-0.59

IntoConstraints

-0.57

liferay

-0.57

__":

-0.56

 Савезне

-0.55

LOAT

-0.55

POSITIVE LOGITS

‎

0.85

‎

0.66

‏

0.63

 Himmel

0.60

‏‏

0.54

‎‎

0.53

.‏

0.52

,‎

0.52

 tantôt

0.47

 vechi

0.47

Activations Density 0.020%

metadata or identifiers associated with document formatting or references

No Comments

No Known Activations