INDEX

Explanations

import statements from Java package declarations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 itſelf

-1.37

 myſelf

-1.34

 Monfieur

-1.15

Vidite

-1.14

хьтан

-1.14

 transfieras

-1.13

 iſt

-1.11

 ―――――

-1.11

 raiſ

-1.07

 Theſe

-1.06

POSITIVE LOGITS

com

1.59

com

0.93

Com

0.70

Com

0.69

0.65

COM

0.61

con

0.60

0.57

COM

0.54

↵↵

0.54

Activations Density 0.013%

import statements from Java package declarations

No Comments

No Known Activations