INDEX

Explanations

references to entries in a dataset or information records

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 kwargs

-0.81

 }}$}

-0.80

"}>

-0.78

FieldBuilder

-0.76

$}

-0.74

"}

-0.74

 Мексичка

-0.73

 nahilalakip

-0.73

❯

-0.73

POSITIVE LOGITS

 entry

1.63

 Entry

1.58

 entries

1.51

entry

1.51

Entries

1.49

 Entries

1.47

Entry

1.46

 ENTRY

1.46

entries

1.42

ENTRY

1.38

Activations Density 0.049%

references to entries in a dataset or information records

No Comments

No Known Activations