INDEX

Explanations

technical terms related to computer architecture

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

Parameters related to memory or cache

np_acts-logits-general · gemini-2.0-flash

hash table lookup

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/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.24.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

 transfection

-0.55

 allAfrica

-0.51

rictions

-0.49

hoeddwyd

-0.49

 umbi

-0.49

saraba

-0.49

hala

-0.48

OURCES

-0.47

rungsseite

-0.47

PyExc

-0.47

POSITIVE LOGITS

 hash

0.98

hash

0.87

 Hash

0.86

 hashing

0.80

Hash

0.78

 bucket

0.76

 entry

0.75

 Entry

0.75

Entry

0.73

key

0.73

Activations Density 0.503%

technical terms related to computer architecture

Parameters related to memory or cache

hash table lookup

No Comments

No Known Activations