INDEX

Explanations

specific identifiers and labels related to data structures or programming elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

co

-0.53

TokenName

-0.51

par

-0.51

bar

-0.47

 tham

-0.47

-0.46

 ко

-0.46

 ",";

-0.46

rix

-0.46

ままで

-0.46

POSITIVE LOGITS

oredCriteria

0.75

adget

0.74

 nakalista

0.70

AndEndTag

0.66

SharedCtor

0.65

webElement

0.65

twimg

0.65

tvguidetime

0.64

WebVitals

0.64

AnimationsModule

0.63

Activations Density 0.371%

specific identifiers and labels related to data structures or programming elements

No Comments

No Known Activations