INDEX

Explanations

references to various studies and their associated results or factors in scientific contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

tagHelperRunner

-0.84

LookAnd

-0.73

UrlResolution

-0.57

 <=",

-0.57

 wireType

-0.56

UnusedPrivate

-0.56

RuleContext

-0.54

GTCX

-0.53

InstrumentedTest

-0.52

 mergeFrom

-0.52

POSITIVE LOGITS

सन्दर्भ

0.42

class

0.41

 espèce

0.40

 Hesse

0.39

CF

0.39

แดง

0.38

0.37

 Hirst

0.36

 Cotter

0.36

sihan

0.36

Activations Density 0.225%

references to various studies and their associated results or factors in scientific contexts

No Comments

No Known Activations