INDEX

Explanations

indicators of the beginning of a new section or document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.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

 Distribute

-0.53

 handling

-0.49

conectar

-0.48

 pistolet

-0.48

vaient

-0.48

 concentra

-0.47

 payable

-0.47

kê

-0.47

lju

-0.47

Palestine

-0.47

POSITIVE LOGITS

 change

1.24

 Change

1.13

 CHANGE

1.11

 changement

1.09

 changes

1.09

change

1.08

Change

1.08

 changements

1.03

 cambiamento

1.02

CHANGE

1.01

Activations Density 0.061%

indicators of the beginning of a new section or document

No Comments

No Known Activations