INDEX

Explanations

terms related to database queries and statistical functions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 ModelExpression

-0.70

 >=",

-0.63

 Савезне

-0.60

 LABORATOR

-0.59

alamualaikum

-0.57

SourceChecksum

-0.56

 ProtoMessage

-0.56

 détruire

-0.56

TagMode

-0.56

 INSTITUTION

-0.55

POSITIVE LOGITS

common

0.72

 متعلقه

0.72

 crdi

0.68

 tfsi

0.65

 Reſ

0.64

✭✭

0.64

ſelf

0.61

ſy

0.61

 Shakspeare

0.61

tagext

0.60

Activations Density 0.602%

terms related to database queries and statistical functions

No Comments

No Known Activations