INDEX

Explanations

lists and categorizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

ništvo

-0.52

السكان

-0.49

ViewType

-0.48

 onBackPressed

-0.47

 uncin

-0.47

 thirds

-0.46

CloseHandle

-0.46

ntype

-0.45

 halftime

-0.45

 anticipo

-0.45

POSITIVE LOGITS

 list

1.90

 lists

1.63

 lista

1.60

 liste

1.60

 List

1.58

 Liste

1.58

 Lists

1.54

 LIST

1.53

 lijst

1.44

 Lista

1.44

Activations Density 0.198%

lists and categorizations

No Comments

No Known Activations