INDEX

Explanations

terms related to macro structures or entities

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 estekak

-0.60

 resourceCulture

-0.58

AccessorTable

-0.57

 OFDb

-0.56

TSCA

-0.52

RTLR

-0.51

bitField

-0.51

stateProvider

-0.48

Erreferentziak

-0.46

expandindo

-0.46

POSITIVE LOGITS

 macro

0.65

ent

0.65

ent

0.63

 threads

0.60

Macro

0.59

 Macro

0.58

Thread

0.56

pit

0.55

ი

0.54

Threads

0.54

Activations Density 0.277%

terms related to macro structures or entities

No Comments

No Known Activations

terms related to macro structures or entities

No Comments

No Known Activations