INDEX

Explanations

important concepts or elements identified as "key" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

fahrung

-0.63

rungsseite

-0.62

UnknownFields

-0.57

 elettron

-0.57

pshire

-0.57

 getItemId

-0.57

 pregunto

-0.56

CardHeader

-0.55

ParallelGroup

-0.55

willReturn

-0.55

POSITIVE LOGITS

key

2.38

Key

1.79

KEY

1.63

 klucz

1.51

 crucial

1.50

Key

1.50

key

1.46

 ключе

1.43

 important

1.40

 Schlüssel

1.36

Activations Density 0.132%

important concepts or elements identified as "key" in various contexts

No Comments

No Known Activations