R future_lapply with SLURM only running sequential

I'm using this SLURM script to initialize the job:

#SBATCH --nodes=1
#SBATCH --ntasks-per-node=1
#SBATCH --cpus-per-task=11
#SBATCH --time=08:00:00

and an Rscript along these lines:

library

data

model_function <- lots of code

plan(multisession, workers=11)

sim_function <- function(sim_run) {
  
  future_lapply(future.seed = TRUE, vals, function(vv){

res <- model_function()

names(res) <- c(...long list of names...)

    dat <- reshape2::melt(res) %>% filter(L1 != "time") %>% mutate(
      time = Var1,
      Var1 = L1,
      L1 = NULL) %>%
      mutate(age_group = cut(Var2, custom_age_groups)) %>%
      group_by(Var1, age_group, time) %>%
      summarise(value = sum(value)) %>%
      mutate(year = ceiling(time / 365),
             vv=vv)

    filename <- paste0(...,vv,".RDS")
    filename_Rt <- paste0(...,vv,".RDS")
    
    saveRDS(dat,filename)
    saveRDS(just_Rt,filename_Rt)

})

filename <- paste0(...,"_sim_run_",sim_run,".RDS")

consolidate_files(filename)

}

lapply(1:20, function(sim_run) sim_function(sim_run))

I can't seem to get this thing to run parallel.

I've tried lots of different plans:

plan(list(tweak(cluster, workers=1),tweak(multisession, workers=11)))
plan(list(tweak(sequential),tweak(multisession)))
plan(list(tweak(sequential),tweak(multicore)))

I've tried using the future_lapply instead:

future_lapply(1:20, function(sim_run) sim_function(sim_run))

All of these still seem to run sequential despite spinning up parallel sessions, is there some topology I'm not seeing?

The code works fine on my Windows desktop so maybe I'm not appropriately using SLURM?