INDEX

Explanations

structured data manipulation and parsing terms within a JSON context, as well as mathematical expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

SafeMath

-0.47

 drop

-0.46

 hunt

-0.44

drop

-0.42

 extension

-0.41

 protoimpl

-0.40

moveToFirst

-0.39

aun

-0.38

 Drop

-0.38

 extend

-0.37

POSITIVE LOGITS

 itſelf

0.93

 myſelf

0.93

 pleaſure

0.91

 Monfieur

0.85

 purpoſe

0.83

 preſent

0.83

 ſeveral

0.82

Efq

0.81

 Chriftian

0.81

 uſe

0.79

Activations Density 0.748%

structured data manipulation and parsing terms within a JSON context, as well as mathematical expressions

No Comments

No Known Activations