INDEX

Explanations

key phrases or concepts indicating specific locations or points of interest

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

Mac

-0.42

Cur

-0.42

Stu

-0.40

Cor

-0.40

Bly

-0.39

gab

-0.38

elek

-0.38

<%@

-0.38

Mas

-0.38

dem

-0.37

POSITIVE LOGITS

ValueStyle

0.99

SourceChecksum

0.98

Efq

0.92

SharedCtor

0.92

 Monfieur

0.89

 purpoſe

0.86

 whoſe

0.86

 myſelf

0.85

 ſta

0.83

 Jefus

0.83

Activations Density 0.207%

key phrases or concepts indicating specific locations or points of interest

No Comments

No Known Activations