INDEX

Explanations

terms related to processes involving immersion or dipping

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IUrlHelper

-0.51

ckså

-0.50

thums

-0.44

eschön

-0.44

WebpackPlugin

-0.42

 кӀ

-0.42

thoven

-0.41

 AssemblyCompany

-0.40

ofollow

-0.39

argout

-0.39

POSITIVE LOGITS

 mergul

0.59

 submer

0.58

 submerged

0.57

 swim

0.56

 Immersion

0.53

 drown

0.51

 immersion

0.50

 Swim

0.50

Swim

0.48

 drowned

0.48

Activations Density 0.040%

terms related to processes involving immersion or dipping

No Comments

No Known Activations