INDEX

Explanations

terms related to reduction or decreases 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 @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

äsident

-0.69

viewDidLoad

-0.58

Pyx

-0.57

RenderAtEndOf

-0.57

CppCodeGen

-0.52

ToScroll

-0.51

 Humb

-0.50

ipun

-0.50

 iParam

-0.49

 RSVP

-0.49

POSITIVE LOGITS

Reduce

0.96

 REDUCTION

0.91

reduce

0.90

 réduction

0.90

REDU

0.89

Reduction

0.88

 Reduction

0.88

 Reduce

0.87

 reduction

0.83

 Redu

0.82

Activations Density 0.133%

terms related to reduction or decreases in various contexts

No Comments

No Known Activations