INDEX

Explanations

technical terms and parameters related to programming

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

 مشين

-0.72

imp

-0.61

 мәкал

-0.60

 impressed

-0.56

++)

-0.56

 שוליים

-0.56

rides

-0.56

 Ganes

-0.55

Numerology

-0.54

knapp

-0.54

POSITIVE LOGITS

didSet

0.72

 quelcon

0.66

 quelconque

0.66

 nôtre

0.64

 eikä

0.61

 étranger

0.60

 anymore

0.60

SBATCH

0.60

 élevées

0.59

 melainkan

0.58

Activations Density 0.659%

technical terms and parameters related to programming

No Comments

No Known Activations