INDEX

Explanations

references to user queries and instructions

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

...@

-0.41

<sup>

-0.40

彼は

-0.39

<em>

-0.38

-0.37

它

-0.36

 Eesti

-0.35

<u>

-0.35

 peningkatan

-0.34

↵↵

-0.34

POSITIVE LOGITS

 AssemblyCulture

1.13

 wireType

1.12

ValueStyle

1.09

MessageOf

1.05

 متعلقه

1.05

 NSCoder

1.05

 intStringLen

1.00

 Normdatei

1.00

 TextAppearance

0.98

 utafitiHapana

0.98

Activations Density 0.002%

references to user queries and instructions

No Comments

No Known Activations