Upload folder using huggingface_hub

.gitattributes

@@ -395,3 +395,13 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 2026.04.27/14.37.44_train_llm_diffusion_unet_lowdim_adroit-hand-hammer-v1/wandb/run-20260427_143833-mhqz8q1s/run-mhqz8q1s.wandb filter=lfs diff=lfs merge=lfs -text
 2026.04.27/14.43.45_train_llm_diffusion_unet_lowdim_adroit-hand-hammer-v1/wandb/run-20260427_144434-ord5zcjk/run-ord5zcjk.wandb filter=lfs diff=lfs merge=lfs -text
 2026.04.27/14.49.53_train_llm_diffusion_unet_lowdim_adroit-hand-hammer-v1/wandb/run-20260427_145041-15x2i90k/run-15x2i90k.wandb filter=lfs diff=lfs merge=lfs -text
+2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/logs.json.txt filter=lfs diff=lfs merge=lfs -text
+2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083436-99c13gi9/run-99c13gi9.wandb filter=lfs diff=lfs merge=lfs -text
+2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083551-da31rpd6/run-da31rpd6.wandb filter=lfs diff=lfs merge=lfs -text
+2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083749-n45frqzz/run-n45frqzz.wandb filter=lfs diff=lfs merge=lfs -text
+2026.04.29/08.43.27_train_llm_diffusion_unet_lowdim_sweep-v2/logs.json.txt filter=lfs diff=lfs merge=lfs -text
+2026.04.29/08.43.27_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_084415-yx0htb6w/run-yx0htb6w.wandb filter=lfs diff=lfs merge=lfs -text
+2026.04.29/08.57.40_train_llm_diffusion_unet_lowdim_sweep-v2/logs.json.txt filter=lfs diff=lfs merge=lfs -text
+2026.04.29/08.57.40_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_085830-fxlvg2pl/run-fxlvg2pl.wandb filter=lfs diff=lfs merge=lfs -text
+2026.04.29/09.04.44_train_llm_diffusion_unet_lowdim_sweep-v2/logs.json.txt filter=lfs diff=lfs merge=lfs -text
+2026.04.29/09.04.44_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_090532-u2idin8m/run-u2idin8m.wandb filter=lfs diff=lfs merge=lfs -text

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/config.yaml

@@ -0,0 +1,193 @@
+name: train_llm_diffusion_unet_lowdim
+_target_: llmbc.workspace.train_llm_diffusion_unet_lowdim_workspace.TrainLLMDiffusionUnetLowdimWorkspace
+obs_dim: ${task.obs_dim}
+action_dim: ${task.action_dim}
+task_name: ${task.name}
+exp_name: default
+model_name: ${llm.name}
+horizon: 1
+n_obs_steps: 1
+n_action_steps: 1
+n_latency_steps: 0
+past_action_visible: false
+keypoint_visible_rate: 1.0
+obs_as_local_cond: false
+obs_as_global_cond: true
+pred_action_steps_only: false
+llm_orig_expert_feedback: true
+llm_do_sample: false
+policy:
+  _target_: llmbc.policy.llm_diffusion_unet_lowdim_policy.LLMDiffusionUnetLowdimPolicy
+  model:
+    _target_: llmbc.model.diffusion.conditional_unet1d.ConditionalUnet1D
+    input_dim: '${eval: ${task.action_dim} if ${obs_as_local_cond} or ${obs_as_global_cond}
+      else ${task.obs_dim} + ${task.action_dim}}'
+    local_cond_dim: '${eval: ${task.obs_dim} if ${obs_as_local_cond} else None}'
+    global_cond_dim: '${eval: ${task.obs_dim}*${n_obs_steps} if ${obs_as_global_cond}
+      else None}'
+    diffusion_step_embed_dim: 256
+    down_dims:
+    - 256
+    - 512
+    - 1024
+    kernel_size: 5
+    n_groups: 8
+    cond_predict_scale: true
+  noise_scheduler:
+    _target_: diffusers.schedulers.scheduling_ddpm.DDPMScheduler
+    num_train_timesteps: 100
+    beta_start: 0.0001
+    beta_end: 0.02
+    beta_schedule: squaredcos_cap_v2
+    variance_type: fixed_small
+    clip_sample: true
+    prediction_type: epsilon
+  horizon: ${horizon}
+  obs_dim: ${obs_dim}
+  action_dim: ${action_dim}
+  n_action_steps: ${eval:'${n_action_steps}+${n_latency_steps}'}
+  n_obs_steps: ${n_obs_steps}
+  num_inference_steps: 100
+  obs_as_local_cond: ${obs_as_local_cond}
+  obs_as_global_cond: ${obs_as_global_cond}
+  pred_action_steps_only: ${pred_action_steps_only}
+  oa_step_convention: true
+  llm_discriminator:
+    _target_: llmbc.discriminator.llm_ce_discriminator.LLMCEDiscriminator
+    task_id: ${task_name}
+    llm_translator:
+      _target_: llmbc.translator.llm_translator.LLMTranslator
+      cfg: ${llm}
+      obs_dim: ${obs_dim}
+      action_dim: ${action_dim}
+      horizon: ${horizon}
+      n_obs_steps: ${n_obs_steps}
+      n_action_steps: ${eval:'${n_action_steps}+${n_latency_steps}'}
+  loss_dp_weight: 1.0
+  loss_llm_weight: 0.001
+  normalize_llm_loss: true
+  reweight_llm_loss: true
+ema:
+  _target_: llmbc.model.diffusion.ema_model.EMAModel
+  update_after_step: 0
+  inv_gamma: 1.0
+  power: 0.75
+  min_value: 0.0
+  max_value: 0.9999
+dataloader:
+  batch_size: 16
+  num_workers: 0
+  shuffle: true
+  pin_memory: false
+  persistent_workers: false
+val_dataloader:
+  batch_size: 16
+  num_workers: 0
+  shuffle: false
+  pin_memory: false
+  persistent_workers: false
+optimizer:
+  _target_: torch.optim.AdamW
+  lr: 0.0001
+  betas:
+  - 0.95
+  - 0.999
+  eps: 1.0e-08
+  weight_decay: 1.0e-06
+training:
+  device: cuda:0
+  seed: 43
+  debug: false
+  resume: true
+  lr_scheduler: cosine
+  lr_warmup_steps: 500
+  num_epochs: 1001
+  gradient_accumulate_every: 16
+  use_ema: true
+  rollout_every: 5
+  checkpoint_every: 5
+  val_every: 1
+  sample_every: 5
+  max_train_steps: null
+  max_val_steps: null
+  tqdm_interval_sec: 1.0
+logging:
+  project: ${task.name}-training-mix
+  resume: true
+  mode: online
+  name: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
+  tags:
+  - ${name}
+  - ${task_name}
+  - ${exp_name}
+  id: null
+  group: null
+checkpoint:
+  topk:
+    monitor_key: test_success_rate
+    mode: max
+    k: 5
+    format_str: epoch={epoch:04d}-test_success_rate={test_success_rate:.3f}.ckpt
+  save_last_ckpt: true
+  save_last_snapshot: false
+multi_run:
+  run_dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+  wandb_name_base: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
+task:
+  name: sweep-v2
+  obs_dim: 9
+  action_dim: 4
+  env_runner:
+    _target_: llmbc.env_runner.metaworld_lowdim_runner.MetaworldLowdimRunner
+    env_name: llf-metaworld-sweep-v2
+    n_train: 10
+    n_test: 50
+    n_envs: 10
+    max_steps: 30
+    n_obs_steps: ${n_obs_steps}
+    n_action_steps: ${n_action_steps}
+    instruction_type: b
+    feedback_type:
+    - hp
+    - hn
+    - fp
+    visual: false
+    discount: 0.9
+  dataset:
+    _target_: llmbc.dataset.metaworld_lowdim_dataset.MetaworldLowdimDataset
+    data_path: datasets/sweep-v2-mix(csp10).pt
+    data_path2: datasets/sweep-v2.pt
+    horizon: ${horizon}
+    pad_before: ${eval:'${n_obs_steps}-1'}
+    pad_after: ${eval:'${n_action_steps}-1'}
+    obs_eef_target: true
+    use_manual_normalizer: false
+    val_ratio: 0.1
+    dummy_normalizer: false
+  instructor:
+    _target_: llmbc.translator.instructor.metaworld_instructor.sweep_v2_instructor.SweepV2Instructor
+llm:
+  name: HuggingFaceTB/SmolLM2-135M-Instruct
+  model_name: SmolLM2-135M-Instruct
+  config_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig
+  causal_lm_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM
+  use_quantization: false
+  use_joint_mlp_projector: true
+  llm_mode: ete-finetuned
+  finetune_mode: orig
+  checkpoint: data/outputs/2025.11.01/23.03.01_train_llm_lowdim_sweep-v2/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-sweep-v2/checkpoint-5690
+  max_length: 100
+  lora_config:
+    r: 32
+    lora_alpha: 64
+    lora_dropout: 0.05
+    bias: none
+    task_type: CAUSAL_LM
+  prompter:
+    _target_: llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter
+    use_joint_mlp_projector: true
+  hydra:
+    job:
+      override_dirname: ${model_name}
+    run:
+      dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${model_name}

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/hydra.yaml

@@ -0,0 +1,155 @@
+hydra:
+  run:
+    dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+  sweep:
+    dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+    subdir: ${hydra.job.num}
+  launcher:
+    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
+  sweeper:
+    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
+    max_batch_size: null
+    params: null
+  help:
+    app_name: ${hydra.job.name}
+    header: '${hydra.help.app_name} is powered by Hydra.
+
+      '
+    footer: 'Powered by Hydra (https://hydra.cc)
+
+      Use --hydra-help to view Hydra specific help
+
+      '
+    template: '${hydra.help.header}
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (group=option)
+
+
+      $APP_CONFIG_GROUPS
+
+
+      == Config ==
+
+      Override anything in the config (foo.bar=value)
+
+
+      $CONFIG
+
+
+      ${hydra.help.footer}
+
+      '
+  hydra_help:
+    template: 'Hydra (${hydra.runtime.version})
+
+      See https://hydra.cc for more info.
+
+
+      == Flags ==
+
+      $FLAGS_HELP
+
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
+      to command line)
+
+
+      $HYDRA_CONFIG_GROUPS
+
+
+      Use ''--cfg hydra'' to Show the Hydra config.
+
+      '
+    hydra_help: ???
+  hydra_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][HYDRA] %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+    root:
+      level: INFO
+      handlers:
+      - console
+    loggers:
+      logging_example:
+        level: DEBUG
+    disable_existing_loggers: false
+  job_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+      file:
+        class: logging.FileHandler
+        formatter: simple
+        filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
+    root:
+      level: INFO
+      handlers:
+      - console
+      - file
+    disable_existing_loggers: false
+  env: {}
+  mode: RUN
+  searchpath: []
+  callbacks: {}
+  output_subdir: .hydra
+  overrides:
+    hydra:
+    - hydra.mode=RUN
+    task:
+    - training.seed=43
+  job:
+    name: train
+    chdir: null
+    override_dirname: training.seed=43
+    id: ???
+    num: ???
+    config_name: llmdp_sweep-v2_mix(csp10).yaml
+    env_set: {}
+    env_copy: []
+    config:
+      override_dirname:
+        kv_sep: '='
+        item_sep: ','
+        exclude_keys: []
+  runtime:
+    version: 1.2.0
+    version_base: '1.2'
+    cwd: /workspace/LLM-BC
+    config_sources:
+    - path: hydra.conf
+      schema: pkg
+      provider: hydra
+    - path: /workspace/LLM-BC/config/mix_data
+      schema: file
+      provider: main
+    - path: ''
+      schema: structured
+      provider: schema
+    output_dir: /workspace/LLM-BC/data/outputs/2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2
+    choices:
+      hydra/env: default
+      hydra/callbacks: null
+      hydra/job_logging: default
+      hydra/hydra_logging: default
+      hydra/hydra_help: default
+      hydra/help: default
+      hydra/sweeper: basic
+      hydra/launcher: basic
+      hydra/output: default
+  verbose: false

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/overrides.yaml

@@ -0,0 +1 @@
+- training.seed=43

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/checkpoints/epoch=0425-test_success_rate=0.980.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4883d1f735ba94ce6c03294b3bce6fb979cfe848b63720acc577e8b2694ccb0
+size 1040498994

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/checkpoints/epoch=0465-test_success_rate=0.960.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbaae1468bee60c9d844a934810ea49f105e9f8d68dcd966f084513e4b4740f0
+size 1040498994

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/checkpoints/epoch=0520-test_success_rate=0.980.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2456fe5ea9b0127524493959dfd5dc6dd50728a5c1615518ca7ca83574e498b
+size 1040498994

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/checkpoints/epoch=0535-test_success_rate=0.980.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9619190e43ba71aaccd6a4555d8e96cc7a58afaa80d98bd64c14a3a6fdd03544
+size 1040498994

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/checkpoints/epoch=0600-test_success_rate=0.980.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4cd1c07b574c72c561d1b95e741ccfe3997484dbbf818ffce835d63a5ca647b2
+size 1040498994

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/checkpoints/latest.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7c52cbdee2945ee873fe33cfe98a7c5b3e0e1102b37b026901baea9cd0c1c0f
+size 1040498994

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/logs.json.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6ce4a5cf4ea216ea685b89eade73b7410df1c98da77f86b263b5bd84d79638b
+size 59399773

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/train.log

@@ -0,0 +1,11 @@
+[2026-04-29 08:33:53,418][numexpr.utils][INFO] - Note: NumExpr detected 36 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 16.
+[2026-04-29 08:33:53,418][numexpr.utils][INFO] - NumExpr defaulting to 16 threads.
+[2026-04-29 08:33:56,818][datasets][INFO] - PyTorch version 2.2.2 available.
+[2026-04-29 08:33:56,819][datasets][INFO] - TensorFlow version 2.15.1 available.
+[2026-04-29 08:33:56,820][datasets][INFO] - JAX version 0.4.30 available.
+[2026-04-29 08:33:58,761][llmbc.model.diffusion.conditional_unet1d][INFO] - number of parameters: 6.501479e+07
+[2026-04-29 08:34:03,905][root][INFO] - running build_ext
+[2026-04-29 08:34:03,909][root][INFO] - building 'mujoco_py.cymj' extension
+[2026-04-29 08:34:03,909][root][INFO] - gcc -pthread -B /venv/llm-bc/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /venv/llm-bc/include -I/venv/llm-bc/include -fPIC -O2 -isystem /venv/llm-bc/include -fPIC -I/venv/llm-bc/lib/python3.9/site-packages/mujoco_py -I/root/.mujoco/mujoco210/include -I/venv/llm-bc/lib/python3.9/site-packages/numpy/core/include -I/venv/llm-bc/include/python3.9 -c /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/cymj.c -o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/cymj.o -fopenmp -w
+[2026-04-29 08:34:31,864][root][INFO] - gcc -pthread -B /venv/llm-bc/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /venv/llm-bc/include -I/venv/llm-bc/include -fPIC -O2 -isystem /venv/llm-bc/include -fPIC -I/venv/llm-bc/lib/python3.9/site-packages/mujoco_py -I/root/.mujoco/mujoco210/include -I/venv/llm-bc/lib/python3.9/site-packages/numpy/core/include -I/venv/llm-bc/include/python3.9 -c /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/gl/osmesashim.c -o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/gl/osmesashim.o -fopenmp -w
+[2026-04-29 08:34:32,013][root][INFO] - gcc -pthread -B /venv/llm-bc/compiler_compat -shared -Wl,-rpath,/venv/llm-bc/lib -Wl,-rpath-link,/venv/llm-bc/lib -L/venv/llm-bc/lib -L/venv/llm-bc/lib -Wl,-rpath,/venv/llm-bc/lib -Wl,-rpath-link,/venv/llm-bc/lib -L/venv/llm-bc/lib /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/cymj.o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/gl/osmesashim.o -L/root/.mujoco/mujoco210/bin -Wl,--enable-new-dtags,-rpath,/root/.mujoco/mujoco210/bin -lmujoco210 -lglewosmesa -lOSMesa -lGL -o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/lib.linux-x86_64-cpython-39/mujoco_py/cymj.cpython-39-x86_64-linux-gnu.so -fopenmp

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083436-99c13gi9/files/config.yaml

@@ -0,0 +1,304 @@
+_target_:
+    value: llmbc.workspace.train_llm_diffusion_unet_lowdim_workspace.TrainLLMDiffusionUnetLowdimWorkspace
+_wandb:
+    value:
+        cli_version: 0.18.6
+        m: []
+        python_version: 3.9.25
+        t:
+            "1":
+                - 1
+                - 2
+                - 3
+                - 5
+                - 11
+                - 12
+                - 41
+                - 49
+                - 50
+                - 51
+                - 53
+                - 55
+                - 71
+                - 83
+                - 95
+                - 98
+                - 100
+                - 105
+            "2":
+                - 1
+                - 2
+                - 3
+                - 5
+                - 11
+                - 12
+                - 41
+                - 49
+                - 50
+                - 51
+                - 53
+                - 55
+                - 71
+                - 83
+                - 95
+                - 98
+                - 100
+                - 105
+            "3":
+                - 13
+                - 15
+                - 16
+                - 23
+                - 55
+                - 61
+            "4": 3.9.25
+            "5": 0.18.6
+            "6": 4.47.1
+            "8":
+                - 5
+            "12": 0.18.6
+            "13": linux-x86_64
+action_dim:
+    value: 4
+checkpoint:
+    value:
+        save_last_ckpt: true
+        save_last_snapshot: false
+        topk:
+            format_str: epoch={epoch:04d}-test_success_rate={test_success_rate:.3f}.ckpt
+            k: 5
+            mode: max
+            monitor_key: test_success_rate
+dataloader:
+    value:
+        batch_size: 16
+        num_workers: 0
+        persistent_workers: false
+        pin_memory: false
+        shuffle: true
+ema:
+    value:
+        _target_: llmbc.model.diffusion.ema_model.EMAModel
+        inv_gamma: 1
+        max_value: 0.9999
+        min_value: 0
+        power: 0.75
+        update_after_step: 0
+exp_name:
+    value: default
+horizon:
+    value: 1
+keypoint_visible_rate:
+    value: 1
+llm:
+    value:
+        causal_lm_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM
+        checkpoint: data/outputs/2025.11.01/23.03.01_train_llm_lowdim_sweep-v2/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-sweep-v2/checkpoint-5690
+        config_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig
+        finetune_mode: orig
+        hydra:
+            job:
+                override_dirname: HuggingFaceTB/SmolLM2-135M-Instruct
+            run:
+                dir: data/outputs/2026.04.29/08.33.50_HuggingFaceTB/SmolLM2-135M-Instruct
+        llm_mode: ete-finetuned
+        lora_config:
+            bias: none
+            lora_alpha: 64
+            lora_dropout: 0.05
+            r: 32
+            task_type: CAUSAL_LM
+        max_length: 100
+        model_name: SmolLM2-135M-Instruct
+        name: HuggingFaceTB/SmolLM2-135M-Instruct
+        prompter:
+            _target_: llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter
+            use_joint_mlp_projector: true
+        use_joint_mlp_projector: true
+        use_quantization: false
+llm_do_sample:
+    value: false
+llm_orig_expert_feedback:
+    value: true
+logging:
+    value:
+        group: null
+        id: null
+        mode: online
+        name: 2026.04.29-08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2
+        project: sweep-v2-training-mix
+        resume: true
+        tags:
+            - train_llm_diffusion_unet_lowdim
+            - sweep-v2
+            - default
+model_name:
+    value: HuggingFaceTB/SmolLM2-135M-Instruct
+multi_run:
+    value:
+        run_dir: data/outputs/2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2
+        wandb_name_base: 2026.04.29-08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2
+n_action_steps:
+    value: 1
+n_latency_steps:
+    value: 0
+n_obs_steps:
+    value: 1
+name:
+    value: train_llm_diffusion_unet_lowdim
+obs_as_global_cond:
+    value: true
+obs_as_local_cond:
+    value: false
+obs_dim:
+    value: 9
+optimizer:
+    value:
+        _target_: torch.optim.AdamW
+        betas:
+            - 0.95
+            - 0.999
+        eps: 1e-08
+        lr: 0.0001
+        weight_decay: 1e-06
+output_dir:
+    value: /workspace/LLM-BC/data/outputs/2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2
+past_action_visible:
+    value: false
+policy:
+    value:
+        _target_: llmbc.policy.llm_diffusion_unet_lowdim_policy.LLMDiffusionUnetLowdimPolicy
+        action_dim: 4
+        horizon: 1
+        llm_discriminator:
+            _target_: llmbc.discriminator.llm_ce_discriminator.LLMCEDiscriminator
+            llm_translator:
+                _target_: llmbc.translator.llm_translator.LLMTranslator
+                action_dim: 4
+                cfg:
+                    causal_lm_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM
+                    checkpoint: data/outputs/2025.11.01/23.03.01_train_llm_lowdim_sweep-v2/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-sweep-v2/checkpoint-5690
+                    config_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig
+                    finetune_mode: orig
+                    hydra:
+                        job:
+                            override_dirname: HuggingFaceTB/SmolLM2-135M-Instruct
+                        run:
+                            dir: data/outputs/2026.04.29/08.33.50_HuggingFaceTB/SmolLM2-135M-Instruct
+                    llm_mode: ete-finetuned
+                    lora_config:
+                        bias: none
+                        lora_alpha: 64
+                        lora_dropout: 0.05
+                        r: 32
+                        task_type: CAUSAL_LM
+                    max_length: 100
+                    model_name: SmolLM2-135M-Instruct
+                    name: HuggingFaceTB/SmolLM2-135M-Instruct
+                    prompter:
+                        _target_: llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter
+                        use_joint_mlp_projector: true
+                    use_joint_mlp_projector: true
+                    use_quantization: false
+                horizon: 1
+                n_action_steps: 1
+                n_obs_steps: 1
+                obs_dim: 9
+            task_id: sweep-v2
+        loss_dp_weight: 1
+        loss_llm_weight: 0.001
+        model:
+            _target_: llmbc.model.diffusion.conditional_unet1d.ConditionalUnet1D
+            cond_predict_scale: true
+            diffusion_step_embed_dim: 256
+            down_dims:
+                - 256
+                - 512
+                - 1024
+            global_cond_dim: 9
+            input_dim: 4
+            kernel_size: 5
+            local_cond_dim: null
+            n_groups: 8
+        n_action_steps: 1
+        n_obs_steps: 1
+        noise_scheduler:
+            _target_: diffusers.schedulers.scheduling_ddpm.DDPMScheduler
+            beta_end: 0.02
+            beta_schedule: squaredcos_cap_v2
+            beta_start: 0.0001
+            clip_sample: true
+            num_train_timesteps: 100
+            prediction_type: epsilon
+            variance_type: fixed_small
+        normalize_llm_loss: true
+        num_inference_steps: 100
+        oa_step_convention: true
+        obs_as_global_cond: true
+        obs_as_local_cond: false
+        obs_dim: 9
+        pred_action_steps_only: false
+        reweight_llm_loss: true
+pred_action_steps_only:
+    value: false
+task:
+    value:
+        action_dim: 4
+        dataset:
+            _target_: llmbc.dataset.metaworld_lowdim_dataset.MetaworldLowdimDataset
+            data_path: datasets/sweep-v2-mix(csp10).pt
+            data_path2: datasets/sweep-v2.pt
+            dummy_normalizer: false
+            horizon: 1
+            obs_eef_target: true
+            pad_after: 0
+            pad_before: 0
+            use_manual_normalizer: false
+            val_ratio: 0.1
+        env_runner:
+            _target_: llmbc.env_runner.metaworld_lowdim_runner.MetaworldLowdimRunner
+            discount: 0.9
+            env_name: llf-metaworld-sweep-v2
+            feedback_type:
+                - hp
+                - hn
+                - fp
+            instruction_type: b
+            max_steps: 30
+            n_action_steps: 1
+            n_envs: 10
+            n_obs_steps: 1
+            n_test: 50
+            n_train: 10
+            visual: false
+        instructor:
+            _target_: llmbc.translator.instructor.metaworld_instructor.sweep_v2_instructor.SweepV2Instructor
+        name: sweep-v2
+        obs_dim: 9
+task_name:
+    value: sweep-v2
+training:
+    value:
+        checkpoint_every: 5
+        debug: false
+        device: cuda:0
+        gradient_accumulate_every: 16
+        lr_scheduler: cosine
+        lr_warmup_steps: 500
+        max_train_steps: null
+        max_val_steps: null
+        num_epochs: 1001
+        resume: true
+        rollout_every: 5
+        sample_every: 5
+        seed: 43
+        tqdm_interval_sec: 1
+        use_ema: true
+        val_every: 1
+val_dataloader:
+    value:
+        batch_size: 16
+        num_workers: 0
+        persistent_workers: false
+        pin_memory: false
+        shuffle: false

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083436-99c13gi9/files/output.log

@@ -0,0 +1,3 @@
+Eval MetaworldLowdimRunner 1/6:   0%|                                                                                                                          | 0/30 [00:00<?, ?it/s]/workspace/LLM-BC/llmbc/common/llfbench_util.py:39: UserWarning: Creating a tensor from a list of numpy.ndarrays is extremely slow. Please consider converting the list to a single numpy.ndarray with numpy.array() before converting to a tensor. (Triggered internally at ../torch/csrc/utils/tensor_new.cpp:275.)
+  obs = torch.tensor(obs, dtype=torch.float32).unsqueeze(dim=0).to(device)
+                                                                                                                                                                                      

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083436-99c13gi9/files/wandb-metadata.json

@@ -0,0 +1,56 @@
+{
+  "os": "Linux-5.15.0-140-generic-x86_64-with-glibc2.35",
+  "python": "3.9.25",
+  "startedAt": "2026-04-29T08:34:36.226685Z",
+  "args": [
+    "--config-path",
+    "config/mix_data",
+    "--config-name",
+    "llmdp_sweep-v2_mix(csp10).yaml",
+    "training.seed=43"
+  ],
+  "program": "/workspace/LLM-BC/./train.py",
+  "codePath": "train.py",
+  "git": {
+    "remote": "https://github.com/CHYang25/LLM-BC.git",
+    "commit": "db8568b3c55288dc78a4218dbe04a88527910abb"
+  },
+  "email": "chris920325@gmail.com",
+  "root": "/workspace/LLM-BC/data/outputs/2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2",
+  "host": "d8640274e028",
+  "username": "root",
+  "executable": "/venv/llm-bc/bin/python3",
+  "codePathLocal": "train.py",
+  "cpu_count": 36,
+  "cpu_count_logical": 36,
+  "gpu": "NVIDIA GeForce RTX 4090",
+  "gpu_count": 2,
+  "disk": {
+    "/": {
+      "total": "569083166720",
+      "used": "375285694464"
+    }
+  },
+  "memory": {
+    "total": "135036993536"
+  },
+  "cpu": {
+    "count": 36,
+    "countLogical": 36
+  },
+  "gpu_nvidia": [
+    {
+      "name": "NVIDIA GeForce RTX 4090",
+      "memoryTotal": "25757220864",
+      "cudaCores": 16384,
+      "architecture": "Ada"
+    },
+    {
+      "name": "NVIDIA GeForce RTX 4090",
+      "memoryTotal": "25757220864",
+      "cudaCores": 16384,
+      "architecture": "Ada"
+    }
+  ],
+  "cudaVersion": "12.4"
+}

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083436-99c13gi9/files/wandb-summary.json

@@ -0,0 +1 @@
+{"val_loss":0.010316567495465279,"_step":334333,"test/cumulative_reward":11.02384949301196,"train/success_rate":0.9,"train_action_mse_error":0.0002314533048775047,"_timestamp":1.7775993432009811e+09,"train_loss_llm":0,"train_loss_dp":0.00023620104184374213,"_runtime":147668.617383877,"train/mean_score":9.109262846543364,"global_step":334333,"_wandb":{"runtime":147668},"test/mean_score":8.496490017409638,"epoch":1000,"test/success_rate":0.82,"train/cumulative_reward":12.4359636034395,"lr":5.931879709653032e-13,"train_loss":0.004364427582608797}

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083436-99c13gi9/run-99c13gi9.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4b68ddaeef65a49b33b78cf233ba17322b33dd0efa40e72898fe8a44dc63723
+size 297232412

2026.04.29/08.33.50_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/wandb-resume.json

@@ -0,0 +1 @@
+{"run_id": "99c13gi9"}

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/config.yaml

@@ -0,0 +1,193 @@
+name: train_llm_diffusion_unet_lowdim
+_target_: llmbc.workspace.train_llm_diffusion_unet_lowdim_workspace.TrainLLMDiffusionUnetLowdimWorkspace
+obs_dim: ${task.obs_dim}
+action_dim: ${task.action_dim}
+task_name: ${task.name}
+exp_name: default
+model_name: ${llm.name}
+horizon: 1
+n_obs_steps: 1
+n_action_steps: 1
+n_latency_steps: 0
+past_action_visible: false
+keypoint_visible_rate: 1.0
+obs_as_local_cond: false
+obs_as_global_cond: true
+pred_action_steps_only: false
+llm_orig_expert_feedback: true
+llm_do_sample: false
+policy:
+  _target_: llmbc.policy.llm_diffusion_unet_lowdim_policy.LLMDiffusionUnetLowdimPolicy
+  model:
+    _target_: llmbc.model.diffusion.conditional_unet1d.ConditionalUnet1D
+    input_dim: '${eval: ${task.action_dim} if ${obs_as_local_cond} or ${obs_as_global_cond}
+      else ${task.obs_dim} + ${task.action_dim}}'
+    local_cond_dim: '${eval: ${task.obs_dim} if ${obs_as_local_cond} else None}'
+    global_cond_dim: '${eval: ${task.obs_dim}*${n_obs_steps} if ${obs_as_global_cond}
+      else None}'
+    diffusion_step_embed_dim: 256
+    down_dims:
+    - 256
+    - 512
+    - 1024
+    kernel_size: 5
+    n_groups: 8
+    cond_predict_scale: true
+  noise_scheduler:
+    _target_: diffusers.schedulers.scheduling_ddpm.DDPMScheduler
+    num_train_timesteps: 100
+    beta_start: 0.0001
+    beta_end: 0.02
+    beta_schedule: squaredcos_cap_v2
+    variance_type: fixed_small
+    clip_sample: true
+    prediction_type: epsilon
+  horizon: ${horizon}
+  obs_dim: ${obs_dim}
+  action_dim: ${action_dim}
+  n_action_steps: ${eval:'${n_action_steps}+${n_latency_steps}'}
+  n_obs_steps: ${n_obs_steps}
+  num_inference_steps: 100
+  obs_as_local_cond: ${obs_as_local_cond}
+  obs_as_global_cond: ${obs_as_global_cond}
+  pred_action_steps_only: ${pred_action_steps_only}
+  oa_step_convention: true
+  llm_discriminator:
+    _target_: llmbc.discriminator.llm_ce_discriminator.LLMCEDiscriminator
+    task_id: ${task_name}
+    llm_translator:
+      _target_: llmbc.translator.llm_translator.LLMTranslator
+      cfg: ${llm}
+      obs_dim: ${obs_dim}
+      action_dim: ${action_dim}
+      horizon: ${horizon}
+      n_obs_steps: ${n_obs_steps}
+      n_action_steps: ${eval:'${n_action_steps}+${n_latency_steps}'}
+  loss_dp_weight: 1.0
+  loss_llm_weight: 0.001
+  normalize_llm_loss: true
+  reweight_llm_loss: true
+ema:
+  _target_: llmbc.model.diffusion.ema_model.EMAModel
+  update_after_step: 0
+  inv_gamma: 1.0
+  power: 0.75
+  min_value: 0.0
+  max_value: 0.9999
+dataloader:
+  batch_size: 16
+  num_workers: 0
+  shuffle: true
+  pin_memory: false
+  persistent_workers: false
+val_dataloader:
+  batch_size: 16
+  num_workers: 0
+  shuffle: false
+  pin_memory: false
+  persistent_workers: false
+optimizer:
+  _target_: torch.optim.AdamW
+  lr: 0.0001
+  betas:
+  - 0.95
+  - 0.999
+  eps: 1.0e-08
+  weight_decay: 1.0e-06
+training:
+  device: cuda:0
+  seed: 44
+  debug: false
+  resume: true
+  lr_scheduler: cosine
+  lr_warmup_steps: 500
+  num_epochs: 1001
+  gradient_accumulate_every: 16
+  use_ema: true
+  rollout_every: 5
+  checkpoint_every: 5
+  val_every: 1
+  sample_every: 5
+  max_train_steps: null
+  max_val_steps: null
+  tqdm_interval_sec: 1.0
+logging:
+  project: ${task.name}-training-mix
+  resume: true
+  mode: online
+  name: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
+  tags:
+  - ${name}
+  - ${task_name}
+  - ${exp_name}
+  id: null
+  group: null
+checkpoint:
+  topk:
+    monitor_key: test_success_rate
+    mode: max
+    k: 5
+    format_str: epoch={epoch:04d}-test_success_rate={test_success_rate:.3f}.ckpt
+  save_last_ckpt: true
+  save_last_snapshot: false
+multi_run:
+  run_dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+  wandb_name_base: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
+task:
+  name: sweep-v2
+  obs_dim: 9
+  action_dim: 4
+  env_runner:
+    _target_: llmbc.env_runner.metaworld_lowdim_runner.MetaworldLowdimRunner
+    env_name: llf-metaworld-sweep-v2
+    n_train: 10
+    n_test: 50
+    n_envs: 10
+    max_steps: 30
+    n_obs_steps: ${n_obs_steps}
+    n_action_steps: ${n_action_steps}
+    instruction_type: b
+    feedback_type:
+    - hp
+    - hn
+    - fp
+    visual: false
+    discount: 0.9
+  dataset:
+    _target_: llmbc.dataset.metaworld_lowdim_dataset.MetaworldLowdimDataset
+    data_path: datasets/sweep-v2-mix(csp10).pt
+    data_path2: datasets/sweep-v2.pt
+    horizon: ${horizon}
+    pad_before: ${eval:'${n_obs_steps}-1'}
+    pad_after: ${eval:'${n_action_steps}-1'}
+    obs_eef_target: true
+    use_manual_normalizer: false
+    val_ratio: 0.1
+    dummy_normalizer: false
+  instructor:
+    _target_: llmbc.translator.instructor.metaworld_instructor.sweep_v2_instructor.SweepV2Instructor
+llm:
+  name: HuggingFaceTB/SmolLM2-135M-Instruct
+  model_name: SmolLM2-135M-Instruct
+  config_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig
+  causal_lm_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM
+  use_quantization: false
+  use_joint_mlp_projector: true
+  llm_mode: ete-finetuned
+  finetune_mode: orig
+  checkpoint: data/outputs/2025.11.01/23.03.01_train_llm_lowdim_sweep-v2/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-sweep-v2/checkpoint-5690
+  max_length: 100
+  lora_config:
+    r: 32
+    lora_alpha: 64
+    lora_dropout: 0.05
+    bias: none
+    task_type: CAUSAL_LM
+  prompter:
+    _target_: llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter
+    use_joint_mlp_projector: true
+  hydra:
+    job:
+      override_dirname: ${model_name}
+    run:
+      dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${model_name}

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/hydra.yaml

@@ -0,0 +1,155 @@
+hydra:
+  run:
+    dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+  sweep:
+    dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+    subdir: ${hydra.job.num}
+  launcher:
+    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
+  sweeper:
+    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
+    max_batch_size: null
+    params: null
+  help:
+    app_name: ${hydra.job.name}
+    header: '${hydra.help.app_name} is powered by Hydra.
+
+      '
+    footer: 'Powered by Hydra (https://hydra.cc)
+
+      Use --hydra-help to view Hydra specific help
+
+      '
+    template: '${hydra.help.header}
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (group=option)
+
+
+      $APP_CONFIG_GROUPS
+
+
+      == Config ==
+
+      Override anything in the config (foo.bar=value)
+
+
+      $CONFIG
+
+
+      ${hydra.help.footer}
+
+      '
+  hydra_help:
+    template: 'Hydra (${hydra.runtime.version})
+
+      See https://hydra.cc for more info.
+
+
+      == Flags ==
+
+      $FLAGS_HELP
+
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
+      to command line)
+
+
+      $HYDRA_CONFIG_GROUPS
+
+
+      Use ''--cfg hydra'' to Show the Hydra config.
+
+      '
+    hydra_help: ???
+  hydra_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][HYDRA] %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+    root:
+      level: INFO
+      handlers:
+      - console
+    loggers:
+      logging_example:
+        level: DEBUG
+    disable_existing_loggers: false
+  job_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+      file:
+        class: logging.FileHandler
+        formatter: simple
+        filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
+    root:
+      level: INFO
+      handlers:
+      - console
+      - file
+    disable_existing_loggers: false
+  env: {}
+  mode: RUN
+  searchpath: []
+  callbacks: {}
+  output_subdir: .hydra
+  overrides:
+    hydra:
+    - hydra.mode=RUN
+    task:
+    - training.seed=44
+  job:
+    name: train
+    chdir: null
+    override_dirname: training.seed=44
+    id: ???
+    num: ???
+    config_name: llmdp_sweep-v2_mix(csp10).yaml
+    env_set: {}
+    env_copy: []
+    config:
+      override_dirname:
+        kv_sep: '='
+        item_sep: ','
+        exclude_keys: []
+  runtime:
+    version: 1.2.0
+    version_base: '1.2'
+    cwd: /workspace/LLM-BC
+    config_sources:
+    - path: hydra.conf
+      schema: pkg
+      provider: hydra
+    - path: /workspace/LLM-BC/config/mix_data
+      schema: file
+      provider: main
+    - path: ''
+      schema: structured
+      provider: schema
+    output_dir: /workspace/LLM-BC/data/outputs/2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2
+    choices:
+      hydra/env: default
+      hydra/callbacks: null
+      hydra/job_logging: default
+      hydra/hydra_logging: default
+      hydra/hydra_help: default
+      hydra/help: default
+      hydra/sweeper: basic
+      hydra/launcher: basic
+      hydra/output: default
+  verbose: false

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/overrides.yaml

@@ -0,0 +1 @@
+- training.seed=44

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/logs.json.txt

@@ -0,0 +1,334 @@
+{"train_loss": 1.2388674020767212, "train_loss_dp": 1.2322754859924316, "train_loss_llm": 6.591882705688477, "global_step": 0, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.3442175388336182, "train_loss_dp": 1.3434481620788574, "train_loss_llm": 0.7693281173706055, "global_step": 1, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.3040227890014648, "train_loss_dp": 1.3026940822601318, "train_loss_llm": 1.3286595344543457, "global_step": 2, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.3045568466186523, "train_loss_dp": 1.3037265539169312, "train_loss_llm": 0.8302857875823975, "global_step": 3, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.3829010725021362, "train_loss_dp": 1.3577861785888672, "train_loss_llm": 25.114906311035156, "global_step": 4, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.4155728816986084, "train_loss_dp": 1.4137077331542969, "train_loss_llm": 1.8651561737060547, "global_step": 5, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.3959581851959229, "train_loss_dp": 1.3930513858795166, "train_loss_llm": 2.9068355560302734, "global_step": 6, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.5207334756851196, "train_loss_dp": 1.519765853881836, "train_loss_llm": 0.9676065444946289, "global_step": 7, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.2659562826156616, "train_loss_dp": 1.2645461559295654, "train_loss_llm": 1.4101123809814453, "global_step": 8, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.5243849754333496, "train_loss_dp": 1.523129940032959, "train_loss_llm": 1.255047082901001, "global_step": 9, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.1253689527511597, "train_loss_dp": 1.1235564947128296, "train_loss_llm": 1.8124998807907104, "global_step": 10, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.836875081062317, "train_loss_dp": 1.8362618684768677, "train_loss_llm": 0.6131739616394043, "global_step": 11, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.112700343132019, "train_loss_dp": 1.1121997833251953, "train_loss_llm": 0.5005240440368652, "global_step": 12, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.3141283988952637, "train_loss_dp": 1.3132636547088623, "train_loss_llm": 0.8647284507751465, "global_step": 13, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.0824880599975586, "train_loss_dp": 1.0799466371536255, "train_loss_llm": 2.5414609909057617, "global_step": 14, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.2707302570343018, "train_loss_dp": 1.2657530307769775, "train_loss_llm": 4.97721004486084, "global_step": 15, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.2204726934432983, "train_loss_dp": 1.2186295986175537, "train_loss_llm": 1.8431363105773926, "global_step": 16, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.1068326234817505, "train_loss_dp": 1.1063164472579956, "train_loss_llm": 0.5162099599838257, "global_step": 17, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.2392385005950928, "train_loss_dp": 1.2377345561981201, "train_loss_llm": 1.5040020942687988, "global_step": 18, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.242624044418335, "train_loss_dp": 1.2419416904449463, "train_loss_llm": 0.6823844909667969, "global_step": 19, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.138157606124878, "train_loss_dp": 1.137233018875122, "train_loss_llm": 0.9245715141296387, "global_step": 20, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.4004040956497192, "train_loss_dp": 1.398963212966919, "train_loss_llm": 1.4409370422363281, "global_step": 21, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.318332552909851, "train_loss_dp": 1.3175277709960938, "train_loss_llm": 0.8048315048217773, "global_step": 22, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.2088450193405151, "train_loss_dp": 1.2079399824142456, "train_loss_llm": 0.9050521850585938, "global_step": 23, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.6157349348068237, "train_loss_dp": 1.6152405738830566, "train_loss_llm": 0.4943828582763672, "global_step": 24, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.108720302581787, "train_loss_dp": 1.1081433296203613, "train_loss_llm": 0.5769481658935547, "global_step": 25, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.657572865486145, "train_loss_dp": 1.6568472385406494, "train_loss_llm": 0.725675106048584, "global_step": 26, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.566896677017212, "train_loss_dp": 1.5657176971435547, "train_loss_llm": 1.1790387630462646, "global_step": 27, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.6360225677490234, "train_loss_dp": 1.627196192741394, "train_loss_llm": 8.826374053955078, "global_step": 28, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.5783209800720215, "train_loss_dp": 1.5778642892837524, "train_loss_llm": 0.4567307233810425, "global_step": 29, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.5616095066070557, "train_loss_dp": 1.5587055683135986, "train_loss_llm": 2.9039573669433594, "global_step": 30, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 0.8448933959007263, "train_loss_dp": 0.8447458744049072, "train_loss_llm": 0.1475048065185547, "global_step": 31, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.0205116271972656, "train_loss_dp": 1.0194404125213623, "train_loss_llm": 1.0712642669677734, "global_step": 32, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.2501715421676636, "train_loss_dp": 1.2489936351776123, "train_loss_llm": 1.1778802871704102, "global_step": 33, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.7386176586151123, "train_loss_dp": 1.7355291843414307, "train_loss_llm": 3.0884265899658203, "global_step": 34, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.053229808807373, "train_loss_dp": 1.049795389175415, "train_loss_llm": 3.434361457824707, "global_step": 35, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.1473748683929443, "train_loss_dp": 1.1453187465667725, "train_loss_llm": 2.0561561584472656, "global_step": 36, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.3342711925506592, "train_loss_dp": 1.3333537578582764, "train_loss_llm": 0.9174466133117676, "global_step": 37, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.2094354629516602, "train_loss_dp": 1.208654761314392, "train_loss_llm": 0.7806581258773804, "global_step": 38, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.3324511051177979, "train_loss_dp": 1.3324511051177979, "train_loss_llm": 0.0, "global_step": 39, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 0.9656134247779846, "train_loss_dp": 0.9648867845535278, "train_loss_llm": 0.7266521453857422, "global_step": 40, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.4339081048965454, "train_loss_dp": 1.4327973127365112, "train_loss_llm": 1.1107735633850098, "global_step": 41, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.4383561611175537, "train_loss_dp": 1.4370396137237549, "train_loss_llm": 1.3165712356567383, "global_step": 42, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.2456368207931519, "train_loss_dp": 1.2441602945327759, "train_loss_llm": 1.4765048027038574, "global_step": 43, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.2812676429748535, "train_loss_dp": 1.280168890953064, "train_loss_llm": 1.0987248420715332, "global_step": 44, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.3905893564224243, "train_loss_dp": 1.3894648551940918, "train_loss_llm": 1.1245224475860596, "global_step": 45, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.6257675886154175, "train_loss_dp": 1.6222846508026123, "train_loss_llm": 3.4829673767089844, "global_step": 46, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 0.8930045962333679, "train_loss_dp": 0.8894293308258057, "train_loss_llm": 3.575265884399414, "global_step": 47, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.569298505783081, "train_loss_dp": 1.5683863162994385, "train_loss_llm": 0.9121360778808594, "global_step": 48, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.3024529218673706, "train_loss_dp": 1.3010343313217163, "train_loss_llm": 1.4186288118362427, "global_step": 49, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 0.9522643089294434, "train_loss_dp": 0.9517556428909302, "train_loss_llm": 0.5086450576782227, "global_step": 50, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.1631828546524048, "train_loss_dp": 1.162337064743042, "train_loss_llm": 0.8458480834960938, "global_step": 51, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.2306510210037231, "train_loss_dp": 1.229649305343628, "train_loss_llm": 1.0017471313476562, "global_step": 52, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.0053949356079102, "train_loss_dp": 1.003873348236084, "train_loss_llm": 1.5216455459594727, "global_step": 53, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.4529292583465576, "train_loss_dp": 1.4517543315887451, "train_loss_llm": 1.1748716831207275, "global_step": 54, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.0653715133666992, "train_loss_dp": 1.0649818181991577, "train_loss_llm": 0.3897533416748047, "global_step": 55, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.3443502187728882, "train_loss_dp": 1.3398493528366089, "train_loss_llm": 4.5009074211120605, "global_step": 56, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.4362457990646362, "train_loss_dp": 1.4355272054672241, "train_loss_llm": 0.7185721397399902, "global_step": 57, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.33949875831604, "train_loss_dp": 1.3376734256744385, "train_loss_llm": 1.8253250122070312, "global_step": 58, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.0528810024261475, "train_loss_dp": 1.0522327423095703, "train_loss_llm": 0.6482086181640625, "global_step": 59, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 0.8752689957618713, "train_loss_dp": 0.8746421933174133, "train_loss_llm": 0.626785933971405, "global_step": 60, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.5650197267532349, "train_loss_dp": 1.564497709274292, "train_loss_llm": 0.5220756530761719, "global_step": 61, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.3290226459503174, "train_loss_dp": 1.3268169164657593, "train_loss_llm": 2.205772876739502, "global_step": 62, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.0407973527908325, "train_loss_dp": 1.0390352010726929, "train_loss_llm": 1.762202262878418, "global_step": 63, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.268754482269287, "train_loss_dp": 1.268418312072754, "train_loss_llm": 0.33615994453430176, "global_step": 64, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.1783772706985474, "train_loss_dp": 1.1782500743865967, "train_loss_llm": 0.12724828720092773, "global_step": 65, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.1861451864242554, "train_loss_dp": 1.1858413219451904, "train_loss_llm": 0.3038376569747925, "global_step": 66, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.9010107517242432, "train_loss_dp": 1.9000146389007568, "train_loss_llm": 0.9960908889770508, "global_step": 67, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.5382814407348633, "train_loss_dp": 1.5377585887908936, "train_loss_llm": 0.5227928161621094, "global_step": 68, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.2397849559783936, "train_loss_dp": 1.2387068271636963, "train_loss_llm": 1.0781803131103516, "global_step": 69, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.5664101839065552, "train_loss_dp": 1.5655030012130737, "train_loss_llm": 0.907188892364502, "global_step": 70, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 0.8132384419441223, "train_loss_dp": 0.8122481107711792, "train_loss_llm": 0.9903373718261719, "global_step": 71, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.2832684516906738, "train_loss_dp": 1.2806549072265625, "train_loss_llm": 2.613522529602051, "global_step": 72, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.4660731554031372, "train_loss_dp": 1.462311863899231, "train_loss_llm": 3.7612924575805664, "global_step": 73, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.670693278312683, "train_loss_dp": 1.6698906421661377, "train_loss_llm": 0.8026752471923828, "global_step": 74, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.210265874862671, "train_loss_dp": 1.2099790573120117, "train_loss_llm": 0.286838173866272, "global_step": 75, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 0.991927981376648, "train_loss_dp": 0.9912078380584717, "train_loss_llm": 0.7201385498046875, "global_step": 76, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.358971357345581, "train_loss_dp": 1.3570804595947266, "train_loss_llm": 1.8908424377441406, "global_step": 77, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.1887080669403076, "train_loss_dp": 1.1843137741088867, "train_loss_llm": 4.394281387329102, "global_step": 78, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.3428627252578735, "train_loss_dp": 1.341726541519165, "train_loss_llm": 1.1361594200134277, "global_step": 79, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 0.9852659702301025, "train_loss_dp": 0.982953667640686, "train_loss_llm": 2.3122940063476562, "global_step": 80, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.5844959020614624, "train_loss_dp": 1.5831801891326904, "train_loss_llm": 1.3157501220703125, "global_step": 81, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.1539615392684937, "train_loss_dp": 1.1539615392684937, "train_loss_llm": 0.0, "global_step": 82, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.4022557735443115, "train_loss_dp": 1.4016964435577393, "train_loss_llm": 0.559356689453125, "global_step": 83, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.1545637845993042, "train_loss_dp": 1.1537353992462158, "train_loss_llm": 0.8284015655517578, "global_step": 84, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 0.9741543531417847, "train_loss_dp": 0.9736528396606445, "train_loss_llm": 0.5015220642089844, "global_step": 85, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.6883440017700195, "train_loss_dp": 1.686600685119629, "train_loss_llm": 1.7433433532714844, "global_step": 86, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.3869191408157349, "train_loss_dp": 1.3861849308013916, "train_loss_llm": 0.7341693639755249, "global_step": 87, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.083802580833435, "train_loss_dp": 1.0834014415740967, "train_loss_llm": 0.4011574387550354, "global_step": 88, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.1007407903671265, "train_loss_dp": 1.100039005279541, "train_loss_llm": 0.7018164396286011, "global_step": 89, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.258071780204773, "train_loss_dp": 1.2573144435882568, "train_loss_llm": 0.7573738098144531, "global_step": 90, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.800818920135498, "train_loss_dp": 1.800118327140808, "train_loss_llm": 0.7005767822265625, "global_step": 91, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.4420007467269897, "train_loss_dp": 1.4411556720733643, "train_loss_llm": 0.8450927734375, "global_step": 92, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.4186537265777588, "train_loss_dp": 1.418412446975708, "train_loss_llm": 0.2413043975830078, "global_step": 93, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.4079365730285645, "train_loss_dp": 1.407548427581787, "train_loss_llm": 0.38817596435546875, "global_step": 94, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.9460475444793701, "train_loss_dp": 1.9440900087356567, "train_loss_llm": 1.9575366973876953, "global_step": 95, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.4745219945907593, "train_loss_dp": 1.4720089435577393, "train_loss_llm": 2.513092041015625, "global_step": 96, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.2669206857681274, "train_loss_dp": 1.265136957168579, "train_loss_llm": 1.783712387084961, "global_step": 97, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.146456003189087, "train_loss_dp": 1.1442296504974365, "train_loss_llm": 2.226386785507202, "global_step": 98, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 0.9735116362571716, "train_loss_dp": 0.972518801689148, "train_loss_llm": 0.992863655090332, "global_step": 99, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.7383379936218262, "train_loss_dp": 1.7367786169052124, "train_loss_llm": 1.559362530708313, "global_step": 100, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.2301892042160034, "train_loss_dp": 1.229820728302002, "train_loss_llm": 0.36852550506591797, "global_step": 101, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.0791608095169067, "train_loss_dp": 1.0784564018249512, "train_loss_llm": 0.7044610977172852, "global_step": 102, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.397878646850586, "train_loss_dp": 1.3968199491500854, "train_loss_llm": 1.0587053298950195, "global_step": 103, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.240686058998108, "train_loss_dp": 1.23933744430542, "train_loss_llm": 1.3485565185546875, "global_step": 104, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.2620381116867065, "train_loss_dp": 1.2602649927139282, "train_loss_llm": 1.7731142044067383, "global_step": 105, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.067171573638916, "train_loss_dp": 1.0668649673461914, "train_loss_llm": 0.3065485954284668, "global_step": 106, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.3243896961212158, "train_loss_dp": 1.3234574794769287, "train_loss_llm": 0.9321613311767578, "global_step": 107, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.0538922548294067, "train_loss_dp": 1.0532056093215942, "train_loss_llm": 0.686678409576416, "global_step": 108, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.1288707256317139, "train_loss_dp": 1.1273982524871826, "train_loss_llm": 1.4724845886230469, "global_step": 109, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.3810135126113892, "train_loss_dp": 1.3809113502502441, "train_loss_llm": 0.10215377807617188, "global_step": 110, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.405809998512268, "train_loss_dp": 1.4039547443389893, "train_loss_llm": 1.85526704788208, "global_step": 111, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.1384334564208984, "train_loss_dp": 1.1375644207000732, "train_loss_llm": 0.8690004348754883, "global_step": 112, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.4959896802902222, "train_loss_dp": 1.4913880825042725, "train_loss_llm": 4.601554870605469, "global_step": 113, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.115427017211914, "train_loss_dp": 1.1147327423095703, "train_loss_llm": 0.694310188293457, "global_step": 114, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.471780776977539, "train_loss_dp": 1.47001051902771, "train_loss_llm": 1.7702445983886719, "global_step": 115, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.7403500080108643, "train_loss_dp": 1.7389066219329834, "train_loss_llm": 1.4433493614196777, "global_step": 116, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.4044861793518066, "train_loss_dp": 1.4041846990585327, "train_loss_llm": 0.3014345169067383, "global_step": 117, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.1883952617645264, "train_loss_dp": 1.1847877502441406, "train_loss_llm": 3.6075267791748047, "global_step": 118, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.397672176361084, "train_loss_dp": 1.3925080299377441, "train_loss_llm": 5.164144515991211, "global_step": 119, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.6425764560699463, "train_loss_dp": 1.6411457061767578, "train_loss_llm": 1.4306970834732056, "global_step": 120, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.1665704250335693, "train_loss_dp": 1.1644372940063477, "train_loss_llm": 2.133073329925537, "global_step": 121, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.442475438117981, "train_loss_dp": 1.4402968883514404, "train_loss_llm": 2.178590774536133, "global_step": 122, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.3432563543319702, "train_loss_dp": 1.3422331809997559, "train_loss_llm": 1.0231997966766357, "global_step": 123, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.5999854803085327, "train_loss_dp": 1.5983721017837524, "train_loss_llm": 1.6133387088775635, "global_step": 124, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.8410570621490479, "train_loss_dp": 1.837514042854309, "train_loss_llm": 3.5430545806884766, "global_step": 125, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.4964112043380737, "train_loss_dp": 1.4953627586364746, "train_loss_llm": 1.0483951568603516, "global_step": 126, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.0382572412490845, "train_loss_dp": 1.0359363555908203, "train_loss_llm": 2.3208274841308594, "global_step": 127, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.3925232887268066, "train_loss_dp": 1.3916833400726318, "train_loss_llm": 0.8399925231933594, "global_step": 128, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.1245297193527222, "train_loss_dp": 1.123524785041809, "train_loss_llm": 1.004913330078125, "global_step": 129, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.1351122856140137, "train_loss_dp": 1.1346909999847412, "train_loss_llm": 0.421272337436676, "global_step": 130, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.2864422798156738, "train_loss_dp": 1.2864422798156738, "train_loss_llm": 0.0, "global_step": 131, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 0.9960746169090271, "train_loss_dp": 0.9954396486282349, "train_loss_llm": 0.6349849700927734, "global_step": 132, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.5244301557540894, "train_loss_dp": 1.523934245109558, "train_loss_llm": 0.4958992004394531, "global_step": 133, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.3470388650894165, "train_loss_dp": 1.346142053604126, "train_loss_llm": 0.8968658447265625, "global_step": 134, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.142440676689148, "train_loss_dp": 1.1422964334487915, "train_loss_llm": 0.1442432403564453, "global_step": 135, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 0.9344951510429382, "train_loss_dp": 0.9344457983970642, "train_loss_llm": 0.049335479736328125, "global_step": 136, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.311477541923523, "train_loss_dp": 1.3051283359527588, "train_loss_llm": 6.349224090576172, "global_step": 137, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.4549978971481323, "train_loss_dp": 1.453377604484558, "train_loss_llm": 1.6202850341796875, "global_step": 138, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.1198748350143433, "train_loss_dp": 1.1194406747817993, "train_loss_llm": 0.43415558338165283, "global_step": 139, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.135230302810669, "train_loss_dp": 1.1347548961639404, "train_loss_llm": 0.47536659240722656, "global_step": 140, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.2194401025772095, "train_loss_dp": 1.2170617580413818, "train_loss_llm": 2.378355026245117, "global_step": 141, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 0.9217919707298279, "train_loss_dp": 0.9208834171295166, "train_loss_llm": 0.9085716009140015, "global_step": 142, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.136486291885376, "train_loss_dp": 1.1349189281463623, "train_loss_llm": 1.567413330078125, "global_step": 143, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.0801128149032593, "train_loss_dp": 1.0797677040100098, "train_loss_llm": 0.3450833559036255, "global_step": 144, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 0.9900317788124084, "train_loss_dp": 0.9886875152587891, "train_loss_llm": 1.3442871570587158, "global_step": 145, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.1418815851211548, "train_loss_dp": 1.1412904262542725, "train_loss_llm": 0.5911111831665039, "global_step": 146, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.1239888668060303, "train_loss_dp": 1.1225569248199463, "train_loss_llm": 1.431924819946289, "global_step": 147, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.202366828918457, "train_loss_dp": 1.2003116607666016, "train_loss_llm": 2.0551111698150635, "global_step": 148, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.2030274868011475, "train_loss_dp": 1.2016737461090088, "train_loss_llm": 1.3537216186523438, "global_step": 149, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.3984814882278442, "train_loss_dp": 1.3917144536972046, "train_loss_llm": 6.767082214355469, "global_step": 150, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 0.9864209294319153, "train_loss_dp": 0.9856945276260376, "train_loss_llm": 0.7263822555541992, "global_step": 151, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.4033918380737305, "train_loss_dp": 1.4024749994277954, "train_loss_llm": 0.9167938232421875, "global_step": 152, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.095431923866272, "train_loss_dp": 1.0949585437774658, "train_loss_llm": 0.4733535051345825, "global_step": 153, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 0.9511296153068542, "train_loss_dp": 0.9504530429840088, "train_loss_llm": 0.6765982508659363, "global_step": 154, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.2107429504394531, "train_loss_dp": 1.2104392051696777, "train_loss_llm": 0.30378276109695435, "global_step": 155, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.1577874422073364, "train_loss_dp": 1.15675687789917, "train_loss_llm": 1.0305962562561035, "global_step": 156, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.4007874727249146, "train_loss_dp": 1.3987444639205933, "train_loss_llm": 2.043064594268799, "global_step": 157, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.3777248859405518, "train_loss_dp": 1.3737266063690186, "train_loss_llm": 3.9983038902282715, "global_step": 158, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.0237888097763062, "train_loss_dp": 1.0232511758804321, "train_loss_llm": 0.5375971794128418, "global_step": 159, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.2301875352859497, "train_loss_dp": 1.2268093824386597, "train_loss_llm": 3.3781814575195312, "global_step": 160, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.1690207719802856, "train_loss_dp": 1.1662453413009644, "train_loss_llm": 2.7753982543945312, "global_step": 161, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.1574712991714478, "train_loss_dp": 1.1574712991714478, "train_loss_llm": 0.0, "global_step": 162, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.223042368888855, "train_loss_dp": 1.221956729888916, "train_loss_llm": 1.0856881141662598, "global_step": 163, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.0296891927719116, "train_loss_dp": 1.0282742977142334, "train_loss_llm": 1.4148635864257812, "global_step": 164, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 0.9525166153907776, "train_loss_dp": 0.9519252181053162, "train_loss_llm": 0.5914077758789062, "global_step": 165, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.0718165636062622, "train_loss_dp": 1.0703264474868774, "train_loss_llm": 1.4901447296142578, "global_step": 166, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 0.9783100485801697, "train_loss_dp": 0.9760985374450684, "train_loss_llm": 2.2115116119384766, "global_step": 167, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.1192525625228882, "train_loss_dp": 1.1175626516342163, "train_loss_llm": 1.6898696422576904, "global_step": 168, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.1511489152908325, "train_loss_dp": 1.149807333946228, "train_loss_llm": 1.341639518737793, "global_step": 169, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.0827611684799194, "train_loss_dp": 1.0809656381607056, "train_loss_llm": 1.795522689819336, "global_step": 170, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.7467385530471802, "train_loss_dp": 1.745513677597046, "train_loss_llm": 1.2249259948730469, "global_step": 171, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.17974054813385, "train_loss_dp": 1.1774327754974365, "train_loss_llm": 2.3077683448791504, "global_step": 172, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.7271137237548828, "train_loss_dp": 1.726534128189087, "train_loss_llm": 0.5795516967773438, "global_step": 173, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.0602941513061523, "train_loss_dp": 1.0563955307006836, "train_loss_llm": 3.898571014404297, "global_step": 174, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.0887119770050049, "train_loss_dp": 1.0882422924041748, "train_loss_llm": 0.4696848392486572, "global_step": 175, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.1951638460159302, "train_loss_dp": 1.193314790725708, "train_loss_llm": 1.8490047454833984, "global_step": 176, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.6294522285461426, "train_loss_dp": 1.628934383392334, "train_loss_llm": 0.5178546905517578, "global_step": 177, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.147679090499878, "train_loss_dp": 1.1469160318374634, "train_loss_llm": 0.7630791664123535, "global_step": 178, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.2891361713409424, "train_loss_dp": 1.288069248199463, "train_loss_llm": 1.0669777393341064, "global_step": 179, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.8736897706985474, "train_loss_dp": 1.8709571361541748, "train_loss_llm": 2.7326393127441406, "global_step": 180, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.1996492147445679, "train_loss_dp": 1.1983051300048828, "train_loss_llm": 1.3440589904785156, "global_step": 181, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.177051305770874, "train_loss_dp": 1.177051305770874, "train_loss_llm": 0.0, "global_step": 182, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 0.7242524027824402, "train_loss_dp": 0.7230463027954102, "train_loss_llm": 1.2061166763305664, "global_step": 183, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.2151724100112915, "train_loss_dp": 1.2143428325653076, "train_loss_llm": 0.8295230865478516, "global_step": 184, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.2501025199890137, "train_loss_dp": 1.2491010427474976, "train_loss_llm": 1.0014405250549316, "global_step": 185, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.1343153715133667, "train_loss_dp": 1.1291550397872925, "train_loss_llm": 5.16033935546875, "global_step": 186, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 0.9605101943016052, "train_loss_dp": 0.9589126706123352, "train_loss_llm": 1.5975160598754883, "global_step": 187, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.0226114988327026, "train_loss_dp": 1.0213017463684082, "train_loss_llm": 1.309748649597168, "global_step": 188, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 0.9178253412246704, "train_loss_dp": 0.9151701927185059, "train_loss_llm": 2.655125141143799, "global_step": 189, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.0042310953140259, "train_loss_dp": 1.002968668937683, "train_loss_llm": 1.2624244689941406, "global_step": 190, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 0.6581118702888489, "train_loss_dp": 0.6568156480789185, "train_loss_llm": 1.2962055206298828, "global_step": 191, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 0.7853236198425293, "train_loss_dp": 0.7825818657875061, "train_loss_llm": 2.7417654991149902, "global_step": 192, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.0843703746795654, "train_loss_dp": 1.083986520767212, "train_loss_llm": 0.3838331699371338, "global_step": 193, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.0361926555633545, "train_loss_dp": 1.035136342048645, "train_loss_llm": 1.0563650131225586, "global_step": 194, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.4386019706726074, "train_loss_dp": 1.4377837181091309, "train_loss_llm": 0.818260669708252, "global_step": 195, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.1302552223205566, "train_loss_dp": 1.1299091577529907, "train_loss_llm": 0.34602975845336914, "global_step": 196, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.001301884651184, "train_loss_dp": 1.000491738319397, "train_loss_llm": 0.8101558685302734, "global_step": 197, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 0.9011714458465576, "train_loss_dp": 0.8995151519775391, "train_loss_llm": 1.65631902217865, "global_step": 198, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.0352778434753418, "train_loss_dp": 1.033476710319519, "train_loss_llm": 1.8011374473571777, "global_step": 199, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.0035055875778198, "train_loss_dp": 1.0016999244689941, "train_loss_llm": 1.805656909942627, "global_step": 200, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.173759937286377, "train_loss_dp": 1.1730964183807373, "train_loss_llm": 0.6634653806686401, "global_step": 201, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.0189990997314453, "train_loss_dp": 1.018202543258667, "train_loss_llm": 0.7965726852416992, "global_step": 202, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.1890335083007812, "train_loss_dp": 1.1838366985321045, "train_loss_llm": 5.196804046630859, "global_step": 203, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.039871335029602, "train_loss_dp": 1.0370564460754395, "train_loss_llm": 2.8148880004882812, "global_step": 204, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.178661823272705, "train_loss_dp": 1.1761789321899414, "train_loss_llm": 2.4829001426696777, "global_step": 205, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.1118525266647339, "train_loss_dp": 1.1113088130950928, "train_loss_llm": 0.5437297821044922, "global_step": 206, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.3669025897979736, "train_loss_dp": 1.3669025897979736, "train_loss_llm": 0.0, "global_step": 207, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.1750134229660034, "train_loss_dp": 1.1744359731674194, "train_loss_llm": 0.5774011611938477, "global_step": 208, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.0056936740875244, "train_loss_dp": 1.0053986310958862, "train_loss_llm": 0.29508423805236816, "global_step": 209, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.1668272018432617, "train_loss_dp": 1.1659350395202637, "train_loss_llm": 0.8921959400177002, "global_step": 210, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.4110729694366455, "train_loss_dp": 1.4109841585159302, "train_loss_llm": 0.08882331848144531, "global_step": 211, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.8266933560371399, "train_loss_dp": 0.8255422711372375, "train_loss_llm": 1.1510963439941406, "global_step": 212, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.158111333847046, "train_loss_dp": 1.1572948694229126, "train_loss_llm": 0.816504716873169, "global_step": 213, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.1952803134918213, "train_loss_dp": 1.194082498550415, "train_loss_llm": 1.1977977752685547, "global_step": 214, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.917536735534668, "train_loss_dp": 0.9156641960144043, "train_loss_llm": 1.8725557327270508, "global_step": 215, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.9580981135368347, "train_loss_dp": 0.9566690921783447, "train_loss_llm": 1.4290122985839844, "global_step": 216, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.891786515712738, "train_loss_dp": 0.8913219571113586, "train_loss_llm": 0.4645436406135559, "global_step": 217, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.112402081489563, "train_loss_dp": 1.1108170747756958, "train_loss_llm": 1.5849800109863281, "global_step": 218, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.8663017749786377, "train_loss_dp": 0.8648581504821777, "train_loss_llm": 1.4436423778533936, "global_step": 219, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.1221469640731812, "train_loss_dp": 1.1209053993225098, "train_loss_llm": 1.2415409088134766, "global_step": 220, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.7760478258132935, "train_loss_dp": 0.7746675610542297, "train_loss_llm": 1.3802440166473389, "global_step": 221, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.1690924167633057, "train_loss_dp": 1.1686795949935913, "train_loss_llm": 0.412830114364624, "global_step": 222, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.1120827198028564, "train_loss_dp": 1.1114788055419922, "train_loss_llm": 0.6038742065429688, "global_step": 223, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.1122658252716064, "train_loss_dp": 1.1117057800292969, "train_loss_llm": 0.5599946975708008, "global_step": 224, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.9833661317825317, "train_loss_dp": 0.9821168184280396, "train_loss_llm": 1.2493019104003906, "global_step": 225, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.0453293323516846, "train_loss_dp": 1.04490327835083, "train_loss_llm": 0.4260730743408203, "global_step": 226, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.8943191766738892, "train_loss_dp": 0.891467809677124, "train_loss_llm": 2.8513946533203125, "global_step": 227, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.1114108562469482, "train_loss_dp": 1.1085188388824463, "train_loss_llm": 2.8919687271118164, "global_step": 228, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.6141865253448486, "train_loss_dp": 0.6133884191513062, "train_loss_llm": 0.7981276512145996, "global_step": 229, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.0980831384658813, "train_loss_dp": 1.0976884365081787, "train_loss_llm": 0.39471352100372314, "global_step": 230, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.130492925643921, "train_loss_dp": 1.1300976276397705, "train_loss_llm": 0.3953409194946289, "global_step": 231, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.9105420112609863, "train_loss_dp": 0.9099901914596558, "train_loss_llm": 0.5517990589141846, "global_step": 232, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.1063554286956787, "train_loss_dp": 1.1053496599197388, "train_loss_llm": 1.005815029144287, "global_step": 233, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.956603467464447, "train_loss_dp": 0.954157829284668, "train_loss_llm": 2.445652484893799, "global_step": 234, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.3623346090316772, "train_loss_dp": 1.3599226474761963, "train_loss_llm": 2.411998748779297, "global_step": 235, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.6935862898826599, "train_loss_dp": 0.6930655837059021, "train_loss_llm": 0.5207210779190063, "global_step": 236, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.9292910695075989, "train_loss_dp": 0.9279356002807617, "train_loss_llm": 1.3554916381835938, "global_step": 237, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.171146273612976, "train_loss_dp": 1.170069694519043, "train_loss_llm": 1.0766181945800781, "global_step": 238, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.3187662363052368, "train_loss_dp": 1.3167996406555176, "train_loss_llm": 1.9666132926940918, "global_step": 239, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.4123095273971558, "train_loss_dp": 1.409790277481079, "train_loss_llm": 2.519197463989258, "global_step": 240, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.0543550252914429, "train_loss_dp": 1.0531542301177979, "train_loss_llm": 1.2008113861083984, "global_step": 241, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.049359917640686, "train_loss_dp": 1.0486977100372314, "train_loss_llm": 0.662247896194458, "global_step": 242, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.338232398033142, "train_loss_dp": 1.338232398033142, "train_loss_llm": 0.0, "global_step": 243, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.3680391311645508, "train_loss_dp": 1.366624116897583, "train_loss_llm": 1.4150428771972656, "global_step": 244, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.9139930605888367, "train_loss_dp": 0.9122378826141357, "train_loss_llm": 1.755192756652832, "global_step": 245, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.9375113248825073, "train_loss_dp": 0.9342877864837646, "train_loss_llm": 3.2235145568847656, "global_step": 246, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.8152047991752625, "train_loss_dp": 0.813875675201416, "train_loss_llm": 1.3290989398956299, "global_step": 247, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.9198944568634033, "train_loss_dp": 0.9163600206375122, "train_loss_llm": 3.5344128608703613, "global_step": 248, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.0359647274017334, "train_loss_dp": 1.0336132049560547, "train_loss_llm": 2.3514955043792725, "global_step": 249, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.0262326002120972, "train_loss_dp": 1.026205062866211, "train_loss_llm": 0.027576446533203125, "global_step": 250, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.9936522245407104, "train_loss_dp": 0.9930360913276672, "train_loss_llm": 0.6161308288574219, "global_step": 251, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.8196955919265747, "train_loss_dp": 0.8186774253845215, "train_loss_llm": 1.0181512832641602, "global_step": 252, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.818446695804596, "train_loss_dp": 0.8173128962516785, "train_loss_llm": 1.1337966918945312, "global_step": 253, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.1110507249832153, "train_loss_dp": 1.1091848611831665, "train_loss_llm": 1.8658056259155273, "global_step": 254, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.1969798803329468, "train_loss_dp": 1.196393609046936, "train_loss_llm": 0.5862817764282227, "global_step": 255, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.1921817064285278, "train_loss_dp": 1.1913694143295288, "train_loss_llm": 0.8123292922973633, "global_step": 256, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 0.7036290764808655, "train_loss_dp": 0.7031139135360718, "train_loss_llm": 0.5151605606079102, "global_step": 257, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.1029995679855347, "train_loss_dp": 1.1010249853134155, "train_loss_llm": 1.9745979309082031, "global_step": 258, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 0.9840030074119568, "train_loss_dp": 0.9833230376243591, "train_loss_llm": 0.6799659729003906, "global_step": 259, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 0.7388753890991211, "train_loss_dp": 0.7388753890991211, "train_loss_llm": 0.0, "global_step": 260, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 0.9224936962127686, "train_loss_dp": 0.9200456142425537, "train_loss_llm": 2.4480667114257812, "global_step": 261, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.1513547897338867, "train_loss_dp": 1.1498630046844482, "train_loss_llm": 1.4918174743652344, "global_step": 262, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 0.5965721011161804, "train_loss_dp": 0.59576416015625, "train_loss_llm": 0.8079452514648438, "global_step": 263, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.144292950630188, "train_loss_dp": 1.14174222946167, "train_loss_llm": 2.550673007965088, "global_step": 264, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 0.706194281578064, "train_loss_dp": 0.7056999802589417, "train_loss_llm": 0.4942913055419922, "global_step": 265, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.0771446228027344, "train_loss_dp": 1.075197696685791, "train_loss_llm": 1.9468803405761719, "global_step": 266, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.2268496751785278, "train_loss_dp": 1.2253323793411255, "train_loss_llm": 1.517264723777771, "global_step": 267, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.0723012685775757, "train_loss_dp": 1.0718514919281006, "train_loss_llm": 0.44975948333740234, "global_step": 268, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.0154857635498047, "train_loss_dp": 1.012235403060913, "train_loss_llm": 3.250316619873047, "global_step": 269, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 0.9631349444389343, "train_loss_dp": 0.9626548290252686, "train_loss_llm": 0.4800877571105957, "global_step": 270, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.0593398809432983, "train_loss_dp": 1.0583552122116089, "train_loss_llm": 0.9846293926239014, "global_step": 271, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 0.9703856706619263, "train_loss_dp": 0.9680812358856201, "train_loss_llm": 2.3044140338897705, "global_step": 272, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.0223912000656128, "train_loss_dp": 1.0212609767913818, "train_loss_llm": 1.1302337646484375, "global_step": 273, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.6957029104232788, "train_loss_dp": 0.6950871348381042, "train_loss_llm": 0.6157584190368652, "global_step": 274, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.0375155210494995, "train_loss_dp": 1.034806728363037, "train_loss_llm": 2.7088406085968018, "global_step": 275, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.834666907787323, "train_loss_dp": 0.833699643611908, "train_loss_llm": 0.9672665596008301, "global_step": 276, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.8195911645889282, "train_loss_dp": 0.8191088438034058, "train_loss_llm": 0.48232269287109375, "global_step": 277, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.9824222326278687, "train_loss_dp": 0.9811854362487793, "train_loss_llm": 1.236821174621582, "global_step": 278, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.0881322622299194, "train_loss_dp": 1.0867507457733154, "train_loss_llm": 1.3815407752990723, "global_step": 279, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.7363006472587585, "train_loss_dp": 0.7355679869651794, "train_loss_llm": 0.7326633930206299, "global_step": 280, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.8685497045516968, "train_loss_dp": 0.8676788210868835, "train_loss_llm": 0.8709096908569336, "global_step": 281, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.0963702201843262, "train_loss_dp": 1.0956306457519531, "train_loss_llm": 0.7396026849746704, "global_step": 282, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.0695204734802246, "train_loss_dp": 1.0689547061920166, "train_loss_llm": 0.5657764673233032, "global_step": 283, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.3515467643737793, "train_loss_dp": 1.3497657775878906, "train_loss_llm": 1.7809834480285645, "global_step": 284, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.0780463218688965, "train_loss_dp": 1.0765752792358398, "train_loss_llm": 1.4710168838500977, "global_step": 285, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.02556574344635, "train_loss_dp": 1.0248643159866333, "train_loss_llm": 0.7014394998550415, "global_step": 286, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.0492587089538574, "train_loss_dp": 1.0488719940185547, "train_loss_llm": 0.38677263259887695, "global_step": 287, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.128947138786316, "train_loss_dp": 1.1271917819976807, "train_loss_llm": 1.7553844451904297, "global_step": 288, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.0433176755905151, "train_loss_dp": 1.0422444343566895, "train_loss_llm": 1.0732455253601074, "global_step": 289, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.0376343727111816, "train_loss_dp": 1.0376343727111816, "train_loss_llm": 0.0, "global_step": 290, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.2192332744598389, "train_loss_dp": 1.2182385921478271, "train_loss_llm": 0.9946675300598145, "global_step": 291, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.9796872138977051, "train_loss_dp": 0.9769675731658936, "train_loss_llm": 2.7196407318115234, "global_step": 292, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.8163902759552002, "train_loss_dp": 0.81536865234375, "train_loss_llm": 1.0216177701950073, "global_step": 293, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.9390847682952881, "train_loss_dp": 0.8967978954315186, "train_loss_llm": 42.28687286376953, "global_step": 294, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.0778462886810303, "train_loss_dp": 1.0765454769134521, "train_loss_llm": 1.3008651733398438, "global_step": 295, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.03171968460083, "train_loss_dp": 1.0310401916503906, "train_loss_llm": 0.6795374155044556, "global_step": 296, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.899282693862915, "train_loss_dp": 0.899282693862915, "train_loss_llm": 0.0, "global_step": 297, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.9680736660957336, "train_loss_dp": 0.9657014608383179, "train_loss_llm": 2.372225284576416, "global_step": 298, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.1859251260757446, "train_loss_dp": 1.185170292854309, "train_loss_llm": 0.7548255920410156, "global_step": 299, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.9003415703773499, "train_loss_dp": 0.899675190448761, "train_loss_llm": 0.6663515567779541, "global_step": 300, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.1594992876052856, "train_loss_dp": 1.158750057220459, "train_loss_llm": 0.7492790222167969, "global_step": 301, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.2096930742263794, "train_loss_dp": 1.1951634883880615, "train_loss_llm": 14.52957534790039, "global_step": 302, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.7827754616737366, "train_loss_dp": 0.781986653804779, "train_loss_llm": 0.7888059616088867, "global_step": 303, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.2754685878753662, "train_loss_dp": 1.2740017175674438, "train_loss_llm": 1.4668374061584473, "global_step": 304, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.8588322997093201, "train_loss_dp": 0.8581957817077637, "train_loss_llm": 0.6365127563476562, "global_step": 305, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.1444827318191528, "train_loss_dp": 1.143121600151062, "train_loss_llm": 1.3611679077148438, "global_step": 306, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.059476375579834, "train_loss_dp": 1.058176040649414, "train_loss_llm": 1.3003935813903809, "global_step": 307, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.1027315855026245, "train_loss_dp": 1.102017879486084, "train_loss_llm": 0.713737964630127, "global_step": 308, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.9382288455963135, "train_loss_dp": 0.9373294115066528, "train_loss_llm": 0.8994221687316895, "global_step": 309, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.0022268295288086, "train_loss_dp": 1.001021146774292, "train_loss_llm": 1.2056515216827393, "global_step": 310, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.9776469469070435, "train_loss_dp": 0.9747797846794128, "train_loss_llm": 2.8671722412109375, "global_step": 311, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.034541130065918, "train_loss_dp": 1.0334199666976929, "train_loss_llm": 1.121189832687378, "global_step": 312, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.0297131538391113, "train_loss_dp": 1.0258519649505615, "train_loss_llm": 3.861204147338867, "global_step": 313, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.0681651830673218, "train_loss_dp": 1.0672656297683716, "train_loss_llm": 0.8995151519775391, "global_step": 314, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.1086066961288452, "train_loss_dp": 1.107616662979126, "train_loss_llm": 0.9899783134460449, "global_step": 315, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.0799013376235962, "train_loss_dp": 1.076128363609314, "train_loss_llm": 3.7729721069335938, "global_step": 316, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.0623064041137695, "train_loss_dp": 0.975284993648529, "train_loss_llm": 87.02137756347656, "global_step": 317, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.0481303930282593, "train_loss_dp": 1.0461969375610352, "train_loss_llm": 1.93345308303833, "global_step": 318, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.7532718181610107, "train_loss_dp": 0.7527172565460205, "train_loss_llm": 0.5545456409454346, "global_step": 319, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.3863590955734253, "train_loss_dp": 1.3854074478149414, "train_loss_llm": 0.9516630172729492, "global_step": 320, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.1350114345550537, "train_loss_dp": 1.134497880935669, "train_loss_llm": 0.5136048793792725, "global_step": 321, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.4167976379394531, "train_loss_dp": 1.416335105895996, "train_loss_llm": 0.46259117126464844, "global_step": 322, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.0172756910324097, "train_loss_dp": 1.01527738571167, "train_loss_llm": 1.9982752799987793, "global_step": 323, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.9068878293037415, "train_loss_dp": 0.9059786796569824, "train_loss_llm": 0.9091517925262451, "global_step": 324, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.236008882522583, "train_loss_dp": 1.2351868152618408, "train_loss_llm": 0.8220615386962891, "global_step": 325, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.9236230254173279, "train_loss_dp": 0.9217624068260193, "train_loss_llm": 1.8606085777282715, "global_step": 326, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.0558834075927734, "train_loss_dp": 1.0504271984100342, "train_loss_llm": 5.456230163574219, "global_step": 327, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.1076027154922485, "train_loss_dp": 1.107146978378296, "train_loss_llm": 0.45574951171875, "global_step": 328, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.0450767278671265, "train_loss_dp": 1.0445219278335571, "train_loss_llm": 0.5547962188720703, "global_step": 329, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.2276790142059326, "train_loss_dp": 1.2273931503295898, "train_loss_llm": 0.2859187722206116, "global_step": 330, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.6618244647979736, "train_loss_dp": 0.6610119342803955, "train_loss_llm": 0.8125548362731934, "global_step": 331, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.264814853668213, "train_loss_dp": 1.264038324356079, "train_loss_llm": 0.7765146493911743, "global_step": 332, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.9713806509971619, "train_loss_dp": 0.9690093994140625, "train_loss_llm": 2.371278762817383, "global_step": 333, "epoch": 0, "lr": 4.2000000000000004e-06}

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/train.log

@@ -0,0 +1,11 @@
+[2026-04-29 08:35:07,827][numexpr.utils][INFO] - Note: NumExpr detected 36 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 16.
+[2026-04-29 08:35:07,827][numexpr.utils][INFO] - NumExpr defaulting to 16 threads.
+[2026-04-29 08:35:10,281][datasets][INFO] - PyTorch version 2.2.2 available.
+[2026-04-29 08:35:10,281][datasets][INFO] - TensorFlow version 2.15.1 available.
+[2026-04-29 08:35:10,282][datasets][INFO] - JAX version 0.4.30 available.
+[2026-04-29 08:35:11,944][llmbc.model.diffusion.conditional_unet1d][INFO] - number of parameters: 6.501479e+07
+[2026-04-29 08:35:19,938][root][INFO] - running build_ext
+[2026-04-29 08:35:19,941][root][INFO] - building 'mujoco_py.cymj' extension
+[2026-04-29 08:35:19,941][root][INFO] - gcc -pthread -B /venv/llm-bc/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /venv/llm-bc/include -I/venv/llm-bc/include -fPIC -O2 -isystem /venv/llm-bc/include -fPIC -I/venv/llm-bc/lib/python3.9/site-packages/mujoco_py -I/root/.mujoco/mujoco210/include -I/venv/llm-bc/lib/python3.9/site-packages/numpy/core/include -I/venv/llm-bc/include/python3.9 -c /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/cymj.c -o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/cymj.o -fopenmp -w
+[2026-04-29 08:35:47,482][root][INFO] - gcc -pthread -B /venv/llm-bc/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /venv/llm-bc/include -I/venv/llm-bc/include -fPIC -O2 -isystem /venv/llm-bc/include -fPIC -I/venv/llm-bc/lib/python3.9/site-packages/mujoco_py -I/root/.mujoco/mujoco210/include -I/venv/llm-bc/lib/python3.9/site-packages/numpy/core/include -I/venv/llm-bc/include/python3.9 -c /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/gl/osmesashim.c -o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/gl/osmesashim.o -fopenmp -w
+[2026-04-29 08:35:47,619][root][INFO] - gcc -pthread -B /venv/llm-bc/compiler_compat -shared -Wl,-rpath,/venv/llm-bc/lib -Wl,-rpath-link,/venv/llm-bc/lib -L/venv/llm-bc/lib -L/venv/llm-bc/lib -Wl,-rpath,/venv/llm-bc/lib -Wl,-rpath-link,/venv/llm-bc/lib -L/venv/llm-bc/lib /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/cymj.o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/gl/osmesashim.o -L/root/.mujoco/mujoco210/bin -Wl,--enable-new-dtags,-rpath,/root/.mujoco/mujoco210/bin -lmujoco210 -lglewosmesa -lOSMesa -lGL -o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/lib.linux-x86_64-cpython-39/mujoco_py/cymj.cpython-39-x86_64-linux-gnu.so -fopenmp

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083551-da31rpd6/files/config.yaml

@@ -0,0 +1,304 @@
+_target_:
+    value: llmbc.workspace.train_llm_diffusion_unet_lowdim_workspace.TrainLLMDiffusionUnetLowdimWorkspace
+_wandb:
+    value:
+        cli_version: 0.18.6
+        m: []
+        python_version: 3.9.25
+        t:
+            "1":
+                - 1
+                - 2
+                - 3
+                - 5
+                - 11
+                - 12
+                - 41
+                - 49
+                - 50
+                - 51
+                - 53
+                - 55
+                - 71
+                - 83
+                - 95
+                - 98
+                - 100
+                - 105
+            "2":
+                - 1
+                - 2
+                - 3
+                - 5
+                - 11
+                - 12
+                - 41
+                - 49
+                - 50
+                - 51
+                - 53
+                - 55
+                - 71
+                - 83
+                - 95
+                - 98
+                - 100
+                - 105
+            "3":
+                - 13
+                - 15
+                - 16
+                - 23
+                - 55
+                - 61
+            "4": 3.9.25
+            "5": 0.18.6
+            "6": 4.47.1
+            "8":
+                - 5
+            "12": 0.18.6
+            "13": linux-x86_64
+action_dim:
+    value: 4
+checkpoint:
+    value:
+        save_last_ckpt: true
+        save_last_snapshot: false
+        topk:
+            format_str: epoch={epoch:04d}-test_success_rate={test_success_rate:.3f}.ckpt
+            k: 5
+            mode: max
+            monitor_key: test_success_rate
+dataloader:
+    value:
+        batch_size: 16
+        num_workers: 0
+        persistent_workers: false
+        pin_memory: false
+        shuffle: true
+ema:
+    value:
+        _target_: llmbc.model.diffusion.ema_model.EMAModel
+        inv_gamma: 1
+        max_value: 0.9999
+        min_value: 0
+        power: 0.75
+        update_after_step: 0
+exp_name:
+    value: default
+horizon:
+    value: 1
+keypoint_visible_rate:
+    value: 1
+llm:
+    value:
+        causal_lm_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM
+        checkpoint: data/outputs/2025.11.01/23.03.01_train_llm_lowdim_sweep-v2/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-sweep-v2/checkpoint-5690
+        config_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig
+        finetune_mode: orig
+        hydra:
+            job:
+                override_dirname: HuggingFaceTB/SmolLM2-135M-Instruct
+            run:
+                dir: data/outputs/2026.04.29/08.35.06_HuggingFaceTB/SmolLM2-135M-Instruct
+        llm_mode: ete-finetuned
+        lora_config:
+            bias: none
+            lora_alpha: 64
+            lora_dropout: 0.05
+            r: 32
+            task_type: CAUSAL_LM
+        max_length: 100
+        model_name: SmolLM2-135M-Instruct
+        name: HuggingFaceTB/SmolLM2-135M-Instruct
+        prompter:
+            _target_: llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter
+            use_joint_mlp_projector: true
+        use_joint_mlp_projector: true
+        use_quantization: false
+llm_do_sample:
+    value: false
+llm_orig_expert_feedback:
+    value: true
+logging:
+    value:
+        group: null
+        id: null
+        mode: online
+        name: 2026.04.29-08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2
+        project: sweep-v2-training-mix
+        resume: true
+        tags:
+            - train_llm_diffusion_unet_lowdim
+            - sweep-v2
+            - default
+model_name:
+    value: HuggingFaceTB/SmolLM2-135M-Instruct
+multi_run:
+    value:
+        run_dir: data/outputs/2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2
+        wandb_name_base: 2026.04.29-08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2
+n_action_steps:
+    value: 1
+n_latency_steps:
+    value: 0
+n_obs_steps:
+    value: 1
+name:
+    value: train_llm_diffusion_unet_lowdim
+obs_as_global_cond:
+    value: true
+obs_as_local_cond:
+    value: false
+obs_dim:
+    value: 9
+optimizer:
+    value:
+        _target_: torch.optim.AdamW
+        betas:
+            - 0.95
+            - 0.999
+        eps: 1e-08
+        lr: 0.0001
+        weight_decay: 1e-06
+output_dir:
+    value: /workspace/LLM-BC/data/outputs/2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2
+past_action_visible:
+    value: false
+policy:
+    value:
+        _target_: llmbc.policy.llm_diffusion_unet_lowdim_policy.LLMDiffusionUnetLowdimPolicy
+        action_dim: 4
+        horizon: 1
+        llm_discriminator:
+            _target_: llmbc.discriminator.llm_ce_discriminator.LLMCEDiscriminator
+            llm_translator:
+                _target_: llmbc.translator.llm_translator.LLMTranslator
+                action_dim: 4
+                cfg:
+                    causal_lm_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM
+                    checkpoint: data/outputs/2025.11.01/23.03.01_train_llm_lowdim_sweep-v2/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-sweep-v2/checkpoint-5690
+                    config_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig
+                    finetune_mode: orig
+                    hydra:
+                        job:
+                            override_dirname: HuggingFaceTB/SmolLM2-135M-Instruct
+                        run:
+                            dir: data/outputs/2026.04.29/08.35.06_HuggingFaceTB/SmolLM2-135M-Instruct
+                    llm_mode: ete-finetuned
+                    lora_config:
+                        bias: none
+                        lora_alpha: 64
+                        lora_dropout: 0.05
+                        r: 32
+                        task_type: CAUSAL_LM
+                    max_length: 100
+                    model_name: SmolLM2-135M-Instruct
+                    name: HuggingFaceTB/SmolLM2-135M-Instruct
+                    prompter:
+                        _target_: llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter
+                        use_joint_mlp_projector: true
+                    use_joint_mlp_projector: true
+                    use_quantization: false
+                horizon: 1
+                n_action_steps: 1
+                n_obs_steps: 1
+                obs_dim: 9
+            task_id: sweep-v2
+        loss_dp_weight: 1
+        loss_llm_weight: 0.001
+        model:
+            _target_: llmbc.model.diffusion.conditional_unet1d.ConditionalUnet1D
+            cond_predict_scale: true
+            diffusion_step_embed_dim: 256
+            down_dims:
+                - 256
+                - 512
+                - 1024
+            global_cond_dim: 9
+            input_dim: 4
+            kernel_size: 5
+            local_cond_dim: null
+            n_groups: 8
+        n_action_steps: 1
+        n_obs_steps: 1
+        noise_scheduler:
+            _target_: diffusers.schedulers.scheduling_ddpm.DDPMScheduler
+            beta_end: 0.02
+            beta_schedule: squaredcos_cap_v2
+            beta_start: 0.0001
+            clip_sample: true
+            num_train_timesteps: 100
+            prediction_type: epsilon
+            variance_type: fixed_small
+        normalize_llm_loss: true
+        num_inference_steps: 100
+        oa_step_convention: true
+        obs_as_global_cond: true
+        obs_as_local_cond: false
+        obs_dim: 9
+        pred_action_steps_only: false
+        reweight_llm_loss: true
+pred_action_steps_only:
+    value: false
+task:
+    value:
+        action_dim: 4
+        dataset:
+            _target_: llmbc.dataset.metaworld_lowdim_dataset.MetaworldLowdimDataset
+            data_path: datasets/sweep-v2-mix(csp10).pt
+            data_path2: datasets/sweep-v2.pt
+            dummy_normalizer: false
+            horizon: 1
+            obs_eef_target: true
+            pad_after: 0
+            pad_before: 0
+            use_manual_normalizer: false
+            val_ratio: 0.1
+        env_runner:
+            _target_: llmbc.env_runner.metaworld_lowdim_runner.MetaworldLowdimRunner
+            discount: 0.9
+            env_name: llf-metaworld-sweep-v2
+            feedback_type:
+                - hp
+                - hn
+                - fp
+            instruction_type: b
+            max_steps: 30
+            n_action_steps: 1
+            n_envs: 10
+            n_obs_steps: 1
+            n_test: 50
+            n_train: 10
+            visual: false
+        instructor:
+            _target_: llmbc.translator.instructor.metaworld_instructor.sweep_v2_instructor.SweepV2Instructor
+        name: sweep-v2
+        obs_dim: 9
+task_name:
+    value: sweep-v2
+training:
+    value:
+        checkpoint_every: 5
+        debug: false
+        device: cuda:0
+        gradient_accumulate_every: 16
+        lr_scheduler: cosine
+        lr_warmup_steps: 500
+        max_train_steps: null
+        max_val_steps: null
+        num_epochs: 1001
+        resume: true
+        rollout_every: 5
+        sample_every: 5
+        seed: 44
+        tqdm_interval_sec: 1
+        use_ema: true
+        val_every: 1
+val_dataloader:
+    value:
+        batch_size: 16
+        num_workers: 0
+        persistent_workers: false
+        pin_memory: false
+        shuffle: false

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083551-da31rpd6/files/output.log

@@ -0,0 +1,72 @@
+Traceback (most recent call last):                                                                                                                                                    
+  File "/workspace/LLM-BC/./train.py", line 35, in <module>
+    main()
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/main.py", line 90, in decorated_main
+    _run_hydra(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 389, in _run_hydra
+    _run_app(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 452, in _run_app
+    run_and_report(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 213, in run_and_report
+    return func()
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 453, in <lambda>
+    lambda: hydra.run(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/hydra.py", line 119, in run
+    ret = run_job(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/core/utils.py", line 186, in run_job
+    ret.return_value = task_function(task_cfg)
+  File "/workspace/LLM-BC/./train.py", line 32, in main
+    workspace.run()
+  File "/workspace/LLM-BC/llmbc/workspace/train_llm_diffusion_unet_lowdim_workspace.py", line 252, in run
+    runner_log = env_runner.run(policy)
+  File "/workspace/LLM-BC/llmbc/env_runner/metaworld_lowdim_runner.py", line 113, in run
+    self.env = AsyncVectorEnv(self.env_fns)
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 168, in __init__
+    self._check_observation_spaces()
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 378, in _check_observation_spaces
+    same_spaces, successes = zip(*[pipe.recv() for pipe in self.parent_pipes])
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 378, in <listcomp>
+    same_spaces, successes = zip(*[pipe.recv() for pipe in self.parent_pipes])
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 250, in recv
+    buf = self._recv_bytes()
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 414, in _recv_bytes
+    buf = self._recv(4)
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 379, in _recv
+    chunk = read(handle, remaining)
+KeyboardInterrupt
+Traceback (most recent call last):
+  File "/workspace/LLM-BC/./train.py", line 35, in <module>
+    main()
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/main.py", line 90, in decorated_main
+    _run_hydra(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 389, in _run_hydra
+    _run_app(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 452, in _run_app
+    run_and_report(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 213, in run_and_report
+    return func()
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 453, in <lambda>
+    lambda: hydra.run(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/hydra.py", line 119, in run
+    ret = run_job(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/core/utils.py", line 186, in run_job
+    ret.return_value = task_function(task_cfg)
+  File "/workspace/LLM-BC/./train.py", line 32, in main
+    workspace.run()
+  File "/workspace/LLM-BC/llmbc/workspace/train_llm_diffusion_unet_lowdim_workspace.py", line 252, in run
+    runner_log = env_runner.run(policy)
+  File "/workspace/LLM-BC/llmbc/env_runner/metaworld_lowdim_runner.py", line 113, in run
+    self.env = AsyncVectorEnv(self.env_fns)
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 168, in __init__
+    self._check_observation_spaces()
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 378, in _check_observation_spaces
+    same_spaces, successes = zip(*[pipe.recv() for pipe in self.parent_pipes])
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 378, in <listcomp>
+    same_spaces, successes = zip(*[pipe.recv() for pipe in self.parent_pipes])
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 250, in recv
+    buf = self._recv_bytes()
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 414, in _recv_bytes
+    buf = self._recv(4)
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 379, in _recv
+    chunk = read(handle, remaining)
+KeyboardInterrupt

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083551-da31rpd6/files/requirements.txt

@@ -0,0 +1,299 @@
+llmbc==0.0.0
+aiohappyeyeballs==2.6.1
+antlr4-python3-runtime==4.9.3
+asciitree==0.3.3
+async-timeout==5.0.1
+attrs==25.4.0
+certifi==2025.10.5
+charset-normalizer==3.4.4
+colorama==0.4.6
+dill==0.3.8
+docstring-parser==0.16
+eval_type_backport==0.2.2
+fasteners==0.20
+filelock==3.17.0
+frozenlist==1.8.0
+fsspec==2024.3.1
+gmpy2==2.2.1
+hf-xet==1.1.8
+idna==3.11
+llvmlite==0.39.1
+MarkupSafe==3.0.2
+mdurl==0.1.2
+mkl-service==2.4.0
+mpmath==1.3.0
+msgpack==1.1.1
+networkx==3.2.1
+packaging==25.0
+propcache==0.3.1
+psutil==7.0.0
+pycparser==2.23
+Pygments==2.19.1
+PyOpenGL==3.1.10
+PySocks==1.7.1
+tzdata==2025.3
+xxhash==3.5.0
+pytz==2025.2
+PyYAML==6.0.2
+regex==2025.9.1
+safetensors==0.5.3
+setuptools==80.9.0
+shtab==1.7.2
+six==1.17.0
+tqdm==4.67.1
+tensorboard==2.15.2
+uvloop==0.22.1
+wheel==0.45.1
+zipp==3.21.0
+tensorflow-estimator==2.15.0
+tensorboard-data-server==0.7.2
+google-auth-oauthlib==1.3.1
+tensorflow==2.15.1
+Jinja2==3.1.6
+markdown-it-py==2.2.0
+multiprocess==0.70.15
+numpy==1.23.5
+pillow==11.3.0
+pip==23.3.2
+python-dateutil==2.9.0.post0
+aiosignal==1.4.0
+brotlicffi==1.0.9.2
+hydra-core==1.2.0
+multidict==6.7.0
+pycares==4.10.0
+rich==14.2.0
+sympy==1.14.0
+typeguard==4.4.4
+aiodns==3.5.0
+urllib3==2.5.0
+yarl==1.18.0
+aiohttp==3.12.15
+pyarrow==21.0.0
+pyarrow-hotfix==0.7
+tokenizers==0.21.0
+av==10.0.0
+einops==0.4.1
+Bottleneck==1.4.2
+mkl_fft==1.3.11
+mkl_random==1.2.8
+numba==0.56.4
+numcodecs==0.12.1
+numexpr==2.10.1
+pandas==2.3.3
+zarr==2.12.0
+datasets==2.19.0
+transformers==4.47.1
+accelerate==1.0.1
+peft==0.14.0
+trl==0.11.4
+torch==2.2.2
+tabulate==0.9.0
+sentencepiece==0.2.0
+pyperclip==1.11.0
+PyOpenGL-accelerate==3.1.10
+pydub==0.25.1
+py-cpuinfo==9.0.0
+pure_eval==0.2.3
+ptyprocess==0.7.0
+nvidia-ml-py==13.595.45
+hjson==3.1.0
+gym-notices==0.1.0
+glfw==2.0.0
+Farama-Notifications==0.0.6
+websockets==15.0.1
+wcwidth==0.6.0
+uc-micro-py==1.0.3
+typing-inspection==0.4.2
+triton==2.2.0
+trimesh==4.12.1
+transforms3d==0.4.2
+traitlets==5.14.3
+threadpoolctl==3.6.0
+tenacity==9.1.2
+sniffio==1.3.1
+smmap==5.0.3
+setproctitle==1.3.7
+sentry-sdk==2.58.0
+semantic-version==2.10.0
+scipy==1.13.1
+rpds-py==0.27.1
+python-multipart==0.0.20
+pyparsing==3.3.2
+pynvml==13.0.1
+pydantic_core==2.46.3
+pycryptodomex==3.23.0
+pyasn1==0.6.3
+grpcio==1.80.0
+platformdirs==4.4.0
+pexpect==4.9.0
+parso==0.8.6
+orjson==3.11.5
+opencv-python==4.11.0.86
+stack-data==0.6.3
+omegaconf==2.2.1
+nvidia-nvtx-cu12==12.1.105
+nvidia-nvjitlink-cu12==12.9.86
+nvidia-nccl-cu12==2.19.3
+nvidia-curand-cu12==10.3.2.106
+nvidia-cufft-cu12==11.0.2.54
+nvidia-cuda-runtime-cu12==12.1.105
+nvidia-cuda-nvrtc-cu12==12.1.105
+nvidia-cuda-cupti-cu12==12.1.105
+nvidia-cublas-cu12==12.1.3.1
+ninja==1.13.0
+narwhals==2.20.0
+lxml==6.1.0
+kiwisolver==1.4.7
+joblib==1.5.3
+jiter==0.14.0
+imageio-ffmpeg==0.6.0
+ImageIO==2.37.2
+h5py==3.14.0
+h11==0.16.0
+fonttools==4.60.2
+ffmpy==1.0.0
+fast_kinematics==0.2.2
+executing==2.2.1
+exceptiongroup==1.3.1
+docker-pycreds==0.4.0
+distro==1.9.0
+decorator==5.2.1
+dacite==1.9.2
+cycler==0.12.1
+contourpy==1.3.0
+cloudpickle==3.1.2
+click==8.1.8
+asttokens==3.0.1
+annotated-types==0.7.0
+annotated-doc==0.0.4
+aiofiles==25.1.0
+absl-py==2.3.1
+uvicorn==0.39.0
+tiktoken==0.8.0
+scikit-learn==1.6.1
+sapien==3.0.0b1
+referencing==0.36.2
+pydantic==2.13.3
+pyasn1_modules==0.4.2
+prompt_toolkit==3.0.52
+nvidia-cusparse-cu12==12.1.0.106
+nvidia-cudnn-cu12==8.9.2.26
+nltk==3.9.2
+mujoco==2.3.7
+mdit-py-plugins==0.3.3
+matplotlib-inline==0.2.1
+matplotlib==3.7.5
+linkify-it-py==2.0.3
+jedi==0.19.2
+huggingface-hub==0.26.2
+httpcore==1.0.9
+gymnasium==0.29.1
+google-pasta==0.2.0
+gitdb==4.0.12
+cryptography==47.0.0
+blobfile==3.0.0
+anyio==4.12.1
+llmbc==0.0.0
+tyro==0.9.1
+toppra==0.6.7
+starlette==0.49.3
+seaborn==0.13.2
+rouge_score==0.1.2
+pynndescent==0.6.0
+nvidia-cusolver-cu12==11.4.5.107
+jsonschema-specifications==2025.9.1
+ipython==8.18.1
+httpx==0.28.1
+gymnasium-robotics==1.2.0
+google-auth==2.49.2
+GitPython==3.1.47
+diffusers==0.31.0
+wandb==0.18.6
+umap-learn==0.5.9.post2
+requests-oauthlib==2.0.0
+openai==2.8.1
+mplib==0.1.1
+jsonschema==4.25.1
+gradio_client==0.2.9
+google-genai==1.47.0
+fastapi==0.128.8
+torchvision==0.17.2
+torchaudio==2.2.2
+sentence-transformers==3.2.1
+pytorch-seed==0.2.0
+deepspeed==0.16.1
+bitsandbytes==0.45.0
+altair==6.0.0
+timm==1.0.22
+gradio==3.36.1
+evaluate==0.4.3
+arm_pytorch_utilities==0.5.0
+tensorflow-io-gcs-filesystem==0.37.1
+Markdown==3.9
+pybullet==3.2.6
+wrapt==1.14.2
+tensorflow-probability==0.23.0
+parse==1.19.1
+dm-tree==0.1.8
+termcolor==3.1.0
+requests==2.32.0
+cffi==2.0.0
+opt_einsum==3.4.0
+labmaze==1.0.6
+typing_extensions==4.15.0
+importlib-resources==5.13.0
+importlib-metadata==5.2.0
+FLAML==2.3.6
+dm-env==1.6
+diskcache==5.6.3
+Cython==0.29.37
+mujoco-py==2.1.2.14
+jaxlib==0.4.30
+dm-control==1.0.14
+cmudict==1.0.13
+syllables==1.0.9
+metaworld==2.0.0
+jax==0.4.30
+highway-env==1.9.1
+gym_bandits==0.0.2
+d4rl==1.1
+pyautogen==0.1.0
+protobuf==4.25.9
+tf-agents==0.19.0
+oauthlib==3.3.1
+pytorch-kinematics==0.7.5
+pygame==2.6.1
+gin-config==0.5.0
+mani_skill==3.0.0b20
+tifffile==2024.8.30
+Shapely==1.8.4
+PyWavelets==1.6.0
+Werkzeug==3.1.8
+astunparse==1.6.3
+scikit-image==0.19.3
+pymunk==6.2.1
+ml-dtypes==0.3.2
+gast==0.7.0
+stable-baselines3==2.2.1
+keras==2.15.0
+libclang==18.1.1
+flatbuffers==25.12.19
+gym==0.23.1
+mjrl==1.0.0
+llfbench==0.1.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.0.0
+inflect==7.3.1
+jaraco.collections==5.1.0
+jaraco.context==5.3.0
+jaraco.functools==4.0.1
+jaraco.text==3.12.1
+more-itertools==10.3.0
+packaging==24.2
+platformdirs==4.2.2
+tomli==2.0.1
+typeguard==4.3.0
+typing_extensions==4.12.2
+wheel==0.45.1
+zipp==3.19.2

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083551-da31rpd6/files/wandb-metadata.json

@@ -0,0 +1,56 @@
+{
+  "os": "Linux-5.15.0-140-generic-x86_64-with-glibc2.35",
+  "python": "3.9.25",
+  "startedAt": "2026-04-29T08:35:51.642706Z",
+  "args": [
+    "--config-path",
+    "config/mix_data",
+    "--config-name",
+    "llmdp_sweep-v2_mix(csp10).yaml",
+    "training.seed=44"
+  ],
+  "program": "/workspace/LLM-BC/./train.py",
+  "codePath": "train.py",
+  "git": {
+    "remote": "https://github.com/CHYang25/LLM-BC.git",
+    "commit": "284b3dcd20dcb4b4fca100fc3721ccc07e7bb4ea"
+  },
+  "email": "chris920325@gmail.com",
+  "root": "/workspace/LLM-BC/data/outputs/2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2",
+  "host": "d8640274e028",
+  "username": "root",
+  "executable": "/venv/llm-bc/bin/python3",
+  "codePathLocal": "train.py",
+  "cpu_count": 36,
+  "cpu_count_logical": 36,
+  "gpu": "NVIDIA GeForce RTX 4090",
+  "gpu_count": 2,
+  "disk": {
+    "/": {
+      "total": "569083166720",
+      "used": "375286108160"
+    }
+  },
+  "memory": {
+    "total": "135036993536"
+  },
+  "cpu": {
+    "count": 36,
+    "countLogical": 36
+  },
+  "gpu_nvidia": [
+    {
+      "name": "NVIDIA GeForce RTX 4090",
+      "memoryTotal": "25757220864",
+      "cudaCores": 16384,
+      "architecture": "Ada"
+    },
+    {
+      "name": "NVIDIA GeForce RTX 4090",
+      "memoryTotal": "25757220864",
+      "cudaCores": 16384,
+      "architecture": "Ada"
+    }
+  ],
+  "cudaVersion": "12.4"
+}

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083551-da31rpd6/files/wandb-summary.json

@@ -0,0 +1 @@
+{"train_loss_llm":2.371278762817383,"epoch":0,"_runtime":274.671002315,"_step":333,"_timestamp":1.777451819259317e+09,"global_step":333,"_wandb":{"runtime":274},"train_loss":0.9713806509971619,"train_loss_dp":0.9690093994140625,"lr":4.2000000000000004e-06}

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083551-da31rpd6/run-da31rpd6.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96f9b9902cf5c59c350df6fb48400ac81c1f1b5a450c2da752e1a669d9c538f7
+size 294912

2026.04.29/08.35.06_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/wandb-resume.json

@@ -0,0 +1 @@
+{"run_id": "da31rpd6"}

2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/config.yaml

@@ -0,0 +1,193 @@
+name: train_llm_diffusion_unet_lowdim
+_target_: llmbc.workspace.train_llm_diffusion_unet_lowdim_workspace.TrainLLMDiffusionUnetLowdimWorkspace
+obs_dim: ${task.obs_dim}
+action_dim: ${task.action_dim}
+task_name: ${task.name}
+exp_name: default
+model_name: ${llm.name}
+horizon: 1
+n_obs_steps: 1
+n_action_steps: 1
+n_latency_steps: 0
+past_action_visible: false
+keypoint_visible_rate: 1.0
+obs_as_local_cond: false
+obs_as_global_cond: true
+pred_action_steps_only: false
+llm_orig_expert_feedback: true
+llm_do_sample: false
+policy:
+  _target_: llmbc.policy.llm_diffusion_unet_lowdim_policy.LLMDiffusionUnetLowdimPolicy
+  model:
+    _target_: llmbc.model.diffusion.conditional_unet1d.ConditionalUnet1D
+    input_dim: '${eval: ${task.action_dim} if ${obs_as_local_cond} or ${obs_as_global_cond}
+      else ${task.obs_dim} + ${task.action_dim}}'
+    local_cond_dim: '${eval: ${task.obs_dim} if ${obs_as_local_cond} else None}'
+    global_cond_dim: '${eval: ${task.obs_dim}*${n_obs_steps} if ${obs_as_global_cond}
+      else None}'
+    diffusion_step_embed_dim: 256
+    down_dims:
+    - 256
+    - 512
+    - 1024
+    kernel_size: 5
+    n_groups: 8
+    cond_predict_scale: true
+  noise_scheduler:
+    _target_: diffusers.schedulers.scheduling_ddpm.DDPMScheduler
+    num_train_timesteps: 100
+    beta_start: 0.0001
+    beta_end: 0.02
+    beta_schedule: squaredcos_cap_v2
+    variance_type: fixed_small
+    clip_sample: true
+    prediction_type: epsilon
+  horizon: ${horizon}
+  obs_dim: ${obs_dim}
+  action_dim: ${action_dim}
+  n_action_steps: ${eval:'${n_action_steps}+${n_latency_steps}'}
+  n_obs_steps: ${n_obs_steps}
+  num_inference_steps: 100
+  obs_as_local_cond: ${obs_as_local_cond}
+  obs_as_global_cond: ${obs_as_global_cond}
+  pred_action_steps_only: ${pred_action_steps_only}
+  oa_step_convention: true
+  llm_discriminator:
+    _target_: llmbc.discriminator.llm_ce_discriminator.LLMCEDiscriminator
+    task_id: ${task_name}
+    llm_translator:
+      _target_: llmbc.translator.llm_translator.LLMTranslator
+      cfg: ${llm}
+      obs_dim: ${obs_dim}
+      action_dim: ${action_dim}
+      horizon: ${horizon}
+      n_obs_steps: ${n_obs_steps}
+      n_action_steps: ${eval:'${n_action_steps}+${n_latency_steps}'}
+  loss_dp_weight: 1.0
+  loss_llm_weight: 0.001
+  normalize_llm_loss: true
+  reweight_llm_loss: true
+ema:
+  _target_: llmbc.model.diffusion.ema_model.EMAModel
+  update_after_step: 0
+  inv_gamma: 1.0
+  power: 0.75
+  min_value: 0.0
+  max_value: 0.9999
+dataloader:
+  batch_size: 16
+  num_workers: 0
+  shuffle: true
+  pin_memory: false
+  persistent_workers: false
+val_dataloader:
+  batch_size: 16
+  num_workers: 0
+  shuffle: false
+  pin_memory: false
+  persistent_workers: false
+optimizer:
+  _target_: torch.optim.AdamW
+  lr: 0.0001
+  betas:
+  - 0.95
+  - 0.999
+  eps: 1.0e-08
+  weight_decay: 1.0e-06
+training:
+  device: cuda:0
+  seed: 45
+  debug: false
+  resume: true
+  lr_scheduler: cosine
+  lr_warmup_steps: 500
+  num_epochs: 1001
+  gradient_accumulate_every: 16
+  use_ema: true
+  rollout_every: 5
+  checkpoint_every: 5
+  val_every: 1
+  sample_every: 5
+  max_train_steps: null
+  max_val_steps: null
+  tqdm_interval_sec: 1.0
+logging:
+  project: ${task.name}-training-mix
+  resume: true
+  mode: online
+  name: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
+  tags:
+  - ${name}
+  - ${task_name}
+  - ${exp_name}
+  id: null
+  group: null
+checkpoint:
+  topk:
+    monitor_key: test_success_rate
+    mode: max
+    k: 5
+    format_str: epoch={epoch:04d}-test_success_rate={test_success_rate:.3f}.ckpt
+  save_last_ckpt: true
+  save_last_snapshot: false
+multi_run:
+  run_dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+  wandb_name_base: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
+task:
+  name: sweep-v2
+  obs_dim: 9
+  action_dim: 4
+  env_runner:
+    _target_: llmbc.env_runner.metaworld_lowdim_runner.MetaworldLowdimRunner
+    env_name: llf-metaworld-sweep-v2
+    n_train: 10
+    n_test: 50
+    n_envs: 10
+    max_steps: 30
+    n_obs_steps: ${n_obs_steps}
+    n_action_steps: ${n_action_steps}
+    instruction_type: b
+    feedback_type:
+    - hp
+    - hn
+    - fp
+    visual: false
+    discount: 0.9
+  dataset:
+    _target_: llmbc.dataset.metaworld_lowdim_dataset.MetaworldLowdimDataset
+    data_path: datasets/sweep-v2-mix(csp10).pt
+    data_path2: datasets/sweep-v2.pt
+    horizon: ${horizon}
+    pad_before: ${eval:'${n_obs_steps}-1'}
+    pad_after: ${eval:'${n_action_steps}-1'}
+    obs_eef_target: true
+    use_manual_normalizer: false
+    val_ratio: 0.1
+    dummy_normalizer: false
+  instructor:
+    _target_: llmbc.translator.instructor.metaworld_instructor.sweep_v2_instructor.SweepV2Instructor
+llm:
+  name: HuggingFaceTB/SmolLM2-135M-Instruct
+  model_name: SmolLM2-135M-Instruct
+  config_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig
+  causal_lm_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM
+  use_quantization: false
+  use_joint_mlp_projector: true
+  llm_mode: ete-finetuned
+  finetune_mode: orig
+  checkpoint: data/outputs/2025.11.01/23.03.01_train_llm_lowdim_sweep-v2/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-sweep-v2/checkpoint-5690
+  max_length: 100
+  lora_config:
+    r: 32
+    lora_alpha: 64
+    lora_dropout: 0.05
+    bias: none
+    task_type: CAUSAL_LM
+  prompter:
+    _target_: llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter
+    use_joint_mlp_projector: true
+  hydra:
+    job:
+      override_dirname: ${model_name}
+    run:
+      dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${model_name}

2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/hydra.yaml

@@ -0,0 +1,155 @@
+hydra:
+  run:
+    dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+  sweep:
+    dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+    subdir: ${hydra.job.num}
+  launcher:
+    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
+  sweeper:
+    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
+    max_batch_size: null
+    params: null
+  help:
+    app_name: ${hydra.job.name}
+    header: '${hydra.help.app_name} is powered by Hydra.
+
+      '
+    footer: 'Powered by Hydra (https://hydra.cc)
+
+      Use --hydra-help to view Hydra specific help
+
+      '
+    template: '${hydra.help.header}
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (group=option)
+
+
+      $APP_CONFIG_GROUPS
+
+
+      == Config ==
+
+      Override anything in the config (foo.bar=value)
+
+
+      $CONFIG
+
+
+      ${hydra.help.footer}
+
+      '
+  hydra_help:
+    template: 'Hydra (${hydra.runtime.version})
+
+      See https://hydra.cc for more info.
+
+
+      == Flags ==
+
+      $FLAGS_HELP
+
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
+      to command line)
+
+
+      $HYDRA_CONFIG_GROUPS
+
+
+      Use ''--cfg hydra'' to Show the Hydra config.
+
+      '
+    hydra_help: ???
+  hydra_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][HYDRA] %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+    root:
+      level: INFO
+      handlers:
+      - console
+    loggers:
+      logging_example:
+        level: DEBUG
+    disable_existing_loggers: false
+  job_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+      file:
+        class: logging.FileHandler
+        formatter: simple
+        filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
+    root:
+      level: INFO
+      handlers:
+      - console
+      - file
+    disable_existing_loggers: false
+  env: {}
+  mode: RUN
+  searchpath: []
+  callbacks: {}
+  output_subdir: .hydra
+  overrides:
+    hydra:
+    - hydra.mode=RUN
+    task:
+    - training.seed=45
+  job:
+    name: train
+    chdir: null
+    override_dirname: training.seed=45
+    id: ???
+    num: ???
+    config_name: llmdp_sweep-v2_mix(csp10).yaml
+    env_set: {}
+    env_copy: []
+    config:
+      override_dirname:
+        kv_sep: '='
+        item_sep: ','
+        exclude_keys: []
+  runtime:
+    version: 1.2.0
+    version_base: '1.2'
+    cwd: /workspace/LLM-BC
+    config_sources:
+    - path: hydra.conf
+      schema: pkg
+      provider: hydra
+    - path: /workspace/LLM-BC/config/mix_data
+      schema: file
+      provider: main
+    - path: ''
+      schema: structured
+      provider: schema
+    output_dir: /workspace/LLM-BC/data/outputs/2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2
+    choices:
+      hydra/env: default
+      hydra/callbacks: null
+      hydra/job_logging: default
+      hydra/hydra_logging: default
+      hydra/hydra_help: default
+      hydra/help: default
+      hydra/sweeper: basic
+      hydra/launcher: basic
+      hydra/output: default
+  verbose: false

2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/overrides.yaml

@@ -0,0 +1 @@
+- training.seed=45

2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/logs.json.txt

@@ -0,0 +1,338 @@
+{"train_loss": 0.7789667248725891, "train_loss_dp": 0.7784344553947449, "train_loss_llm": 0.532257080078125, "global_step": 0, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.4812785387039185, "train_loss_dp": 1.480205774307251, "train_loss_llm": 1.0727070569992065, "global_step": 1, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.703511118888855, "train_loss_dp": 1.7020270824432373, "train_loss_llm": 1.4839982986450195, "global_step": 2, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 0.9466181397438049, "train_loss_dp": 0.9464600682258606, "train_loss_llm": 0.15809392929077148, "global_step": 3, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.3648229837417603, "train_loss_dp": 1.3631024360656738, "train_loss_llm": 1.7205419540405273, "global_step": 4, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 0.9820842742919922, "train_loss_dp": 0.9814388751983643, "train_loss_llm": 0.6454005241394043, "global_step": 5, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.3251570463180542, "train_loss_dp": 1.324012041091919, "train_loss_llm": 1.1449956893920898, "global_step": 6, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 0.8598686456680298, "train_loss_dp": 0.8584091067314148, "train_loss_llm": 1.4595112800598145, "global_step": 7, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.1753065586090088, "train_loss_dp": 1.1745007038116455, "train_loss_llm": 0.8058884143829346, "global_step": 8, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.299705982208252, "train_loss_dp": 1.2982532978057861, "train_loss_llm": 1.4526424407958984, "global_step": 9, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 0.9544914364814758, "train_loss_dp": 0.9538720846176147, "train_loss_llm": 0.6193599700927734, "global_step": 10, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 0.9061971306800842, "train_loss_dp": 0.9056134223937988, "train_loss_llm": 0.5836952924728394, "global_step": 11, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.3258334398269653, "train_loss_dp": 1.32527494430542, "train_loss_llm": 0.5585168600082397, "global_step": 12, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.3159362077713013, "train_loss_dp": 1.3154011964797974, "train_loss_llm": 0.5349661111831665, "global_step": 13, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.435994267463684, "train_loss_dp": 1.4338123798370361, "train_loss_llm": 2.181854248046875, "global_step": 14, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 1.075801968574524, "train_loss_dp": 1.0663318634033203, "train_loss_llm": 9.470149993896484, "global_step": 15, "epoch": 0, "lr": 2.0000000000000002e-07}
+{"train_loss": 0.9878379702568054, "train_loss_dp": 0.9851472973823547, "train_loss_llm": 2.6906909942626953, "global_step": 16, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.0550199747085571, "train_loss_dp": 1.054060697555542, "train_loss_llm": 0.9592361450195312, "global_step": 17, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 0.9808177351951599, "train_loss_dp": 0.978840708732605, "train_loss_llm": 1.97705078125, "global_step": 18, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 0.8384479880332947, "train_loss_dp": 0.8380396962165833, "train_loss_llm": 0.40830904245376587, "global_step": 19, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.117823839187622, "train_loss_dp": 1.112903118133545, "train_loss_llm": 4.920753002166748, "global_step": 20, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 0.7604645490646362, "train_loss_dp": 0.7604645490646362, "train_loss_llm": 0.0, "global_step": 21, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.1393953561782837, "train_loss_dp": 1.1382240056991577, "train_loss_llm": 1.1713260412216187, "global_step": 22, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.213979721069336, "train_loss_dp": 1.212296485900879, "train_loss_llm": 1.6831917762756348, "global_step": 23, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 0.9646763801574707, "train_loss_dp": 0.9637986421585083, "train_loss_llm": 0.8777172565460205, "global_step": 24, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.5793468952178955, "train_loss_dp": 1.578649878501892, "train_loss_llm": 0.6970024108886719, "global_step": 25, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.2663580179214478, "train_loss_dp": 1.2648851871490479, "train_loss_llm": 1.4728822708129883, "global_step": 26, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.2766700983047485, "train_loss_dp": 1.274940013885498, "train_loss_llm": 1.7301355600357056, "global_step": 27, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.1808645725250244, "train_loss_dp": 1.1778535842895508, "train_loss_llm": 3.0109810829162598, "global_step": 28, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.4790345430374146, "train_loss_dp": 1.4786677360534668, "train_loss_llm": 0.36677098274230957, "global_step": 29, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 0.9326168298721313, "train_loss_dp": 0.9326168298721313, "train_loss_llm": 0.0, "global_step": 30, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.0168713331222534, "train_loss_dp": 1.013830542564392, "train_loss_llm": 3.0407767295837402, "global_step": 31, "epoch": 0, "lr": 4.0000000000000003e-07}
+{"train_loss": 1.0716356039047241, "train_loss_dp": 1.069759488105774, "train_loss_llm": 1.8761253356933594, "global_step": 32, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.3149069547653198, "train_loss_dp": 1.3071506023406982, "train_loss_llm": 7.756320953369141, "global_step": 33, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.3070275783538818, "train_loss_dp": 1.3061859607696533, "train_loss_llm": 0.8416301012039185, "global_step": 34, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.010388970375061, "train_loss_dp": 1.0085391998291016, "train_loss_llm": 1.8497505187988281, "global_step": 35, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.0067718029022217, "train_loss_dp": 1.0066795349121094, "train_loss_llm": 0.0922393798828125, "global_step": 36, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.0842759609222412, "train_loss_dp": 1.0812385082244873, "train_loss_llm": 3.0374603271484375, "global_step": 37, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.3997782468795776, "train_loss_dp": 1.3964835405349731, "train_loss_llm": 3.294764518737793, "global_step": 38, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.138811707496643, "train_loss_dp": 1.1378529071807861, "train_loss_llm": 0.9587554931640625, "global_step": 39, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.2324931621551514, "train_loss_dp": 1.23075532913208, "train_loss_llm": 1.737889051437378, "global_step": 40, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.2905280590057373, "train_loss_dp": 1.2852116823196411, "train_loss_llm": 5.316364288330078, "global_step": 41, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.2838133573532104, "train_loss_dp": 1.2828341722488403, "train_loss_llm": 0.9791371822357178, "global_step": 42, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.1660711765289307, "train_loss_dp": 1.165443778038025, "train_loss_llm": 0.627385139465332, "global_step": 43, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.145284652709961, "train_loss_dp": 1.1441025733947754, "train_loss_llm": 1.1820387840270996, "global_step": 44, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 0.9939045310020447, "train_loss_dp": 0.9935083985328674, "train_loss_llm": 0.3961343765258789, "global_step": 45, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.3373174667358398, "train_loss_dp": 1.3364193439483643, "train_loss_llm": 0.898101806640625, "global_step": 46, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 0.9487116932868958, "train_loss_dp": 0.9471842646598816, "train_loss_llm": 1.527456283569336, "global_step": 47, "epoch": 0, "lr": 6.000000000000001e-07}
+{"train_loss": 1.25667405128479, "train_loss_dp": 1.2556705474853516, "train_loss_llm": 1.0035312175750732, "global_step": 48, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.2156152725219727, "train_loss_dp": 1.2142655849456787, "train_loss_llm": 1.349653720855713, "global_step": 49, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.1537072658538818, "train_loss_dp": 1.1532894372940063, "train_loss_llm": 0.417813777923584, "global_step": 50, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.1653023958206177, "train_loss_dp": 1.1607341766357422, "train_loss_llm": 4.568267822265625, "global_step": 51, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 0.8025030493736267, "train_loss_dp": 0.8018094897270203, "train_loss_llm": 0.693543016910553, "global_step": 52, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 0.9360765218734741, "train_loss_dp": 0.9353166818618774, "train_loss_llm": 0.7598371505737305, "global_step": 53, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 0.8400998115539551, "train_loss_dp": 0.8396124243736267, "train_loss_llm": 0.48740577697753906, "global_step": 54, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.26108980178833, "train_loss_dp": 1.2595584392547607, "train_loss_llm": 1.5313596725463867, "global_step": 55, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 0.8736547827720642, "train_loss_dp": 0.8727149963378906, "train_loss_llm": 0.9397706985473633, "global_step": 56, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.042411208152771, "train_loss_dp": 1.0404326915740967, "train_loss_llm": 1.978510856628418, "global_step": 57, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.0089004039764404, "train_loss_dp": 1.007153034210205, "train_loss_llm": 1.7473623752593994, "global_step": 58, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 0.8581745624542236, "train_loss_dp": 0.837749719619751, "train_loss_llm": 20.424816131591797, "global_step": 59, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.2599515914916992, "train_loss_dp": 1.2589569091796875, "train_loss_llm": 0.9947166442871094, "global_step": 60, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.1447588205337524, "train_loss_dp": 1.1410822868347168, "train_loss_llm": 3.676481246948242, "global_step": 61, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 0.9488521814346313, "train_loss_dp": 0.9466419816017151, "train_loss_llm": 2.2102208137512207, "global_step": 62, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.1681071519851685, "train_loss_dp": 1.1661149263381958, "train_loss_llm": 1.9922842979431152, "global_step": 63, "epoch": 0, "lr": 8.000000000000001e-07}
+{"train_loss": 1.4025390148162842, "train_loss_dp": 1.401512622833252, "train_loss_llm": 1.0264418125152588, "global_step": 64, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.1621136665344238, "train_loss_dp": 1.161423683166504, "train_loss_llm": 0.6899653077125549, "global_step": 65, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 0.6015030741691589, "train_loss_dp": 0.6007305383682251, "train_loss_llm": 0.7725367546081543, "global_step": 66, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.5044846534729004, "train_loss_dp": 1.5034410953521729, "train_loss_llm": 1.043562412261963, "global_step": 67, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 0.8388612866401672, "train_loss_dp": 0.8372955322265625, "train_loss_llm": 1.5657382011413574, "global_step": 68, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.7217475175857544, "train_loss_dp": 1.7209349870681763, "train_loss_llm": 0.8125228881835938, "global_step": 69, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 0.7477439641952515, "train_loss_dp": 0.7448316812515259, "train_loss_llm": 2.912308692932129, "global_step": 70, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.0171953439712524, "train_loss_dp": 1.0148653984069824, "train_loss_llm": 2.329975128173828, "global_step": 71, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 0.7033337354660034, "train_loss_dp": 0.7029812932014465, "train_loss_llm": 0.352466344833374, "global_step": 72, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.5708736181259155, "train_loss_dp": 1.5689499378204346, "train_loss_llm": 1.9236726760864258, "global_step": 73, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.171140432357788, "train_loss_dp": 1.1672642230987549, "train_loss_llm": 3.8761959075927734, "global_step": 74, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 0.945916473865509, "train_loss_dp": 0.9446393251419067, "train_loss_llm": 1.277151346206665, "global_step": 75, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.2723770141601562, "train_loss_dp": 1.2694239616394043, "train_loss_llm": 2.9530458450317383, "global_step": 76, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.08467698097229, "train_loss_dp": 1.0837539434432983, "train_loss_llm": 0.9230525493621826, "global_step": 77, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.134382724761963, "train_loss_dp": 1.1334881782531738, "train_loss_llm": 0.8945956230163574, "global_step": 78, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.1263834238052368, "train_loss_dp": 1.1259883642196655, "train_loss_llm": 0.39508819580078125, "global_step": 79, "epoch": 0, "lr": 1.0000000000000002e-06}
+{"train_loss": 1.133946418762207, "train_loss_dp": 1.1318979263305664, "train_loss_llm": 2.0484776496887207, "global_step": 80, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 0.966083288192749, "train_loss_dp": 0.9656722545623779, "train_loss_llm": 0.41105806827545166, "global_step": 81, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 0.8245104551315308, "train_loss_dp": 0.8229804039001465, "train_loss_llm": 1.5300674438476562, "global_step": 82, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 0.9492079019546509, "train_loss_dp": 0.9480666518211365, "train_loss_llm": 1.1412239074707031, "global_step": 83, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.125891923904419, "train_loss_dp": 1.1252224445343018, "train_loss_llm": 0.669519305229187, "global_step": 84, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 0.9792460799217224, "train_loss_dp": 0.9777927398681641, "train_loss_llm": 1.4533462524414062, "global_step": 85, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.3158930540084839, "train_loss_dp": 1.315097451210022, "train_loss_llm": 0.7955795526504517, "global_step": 86, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.2849777936935425, "train_loss_dp": 1.2843527793884277, "train_loss_llm": 0.6250190734863281, "global_step": 87, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.1404515504837036, "train_loss_dp": 1.13894522190094, "train_loss_llm": 1.5063276290893555, "global_step": 88, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 0.9223363995552063, "train_loss_dp": 0.9214972257614136, "train_loss_llm": 0.8391933441162109, "global_step": 89, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.1727333068847656, "train_loss_dp": 1.1718686819076538, "train_loss_llm": 0.8645706176757812, "global_step": 90, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.113487720489502, "train_loss_dp": 1.1125094890594482, "train_loss_llm": 0.9781861305236816, "global_step": 91, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.3552762269973755, "train_loss_dp": 1.354949712753296, "train_loss_llm": 0.3265666961669922, "global_step": 92, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 0.9690829515457153, "train_loss_dp": 0.968060314655304, "train_loss_llm": 1.0226551294326782, "global_step": 93, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 0.9885964393615723, "train_loss_dp": 0.9868841171264648, "train_loss_llm": 1.7123019695281982, "global_step": 94, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.1996806859970093, "train_loss_dp": 1.1987500190734863, "train_loss_llm": 0.9306955337524414, "global_step": 95, "epoch": 0, "lr": 1.2000000000000002e-06}
+{"train_loss": 1.034594178199768, "train_loss_dp": 1.0339114665985107, "train_loss_llm": 0.6826953887939453, "global_step": 96, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.5009411573410034, "train_loss_dp": 1.4976829290390015, "train_loss_llm": 3.258220672607422, "global_step": 97, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 0.6479157209396362, "train_loss_dp": 0.6475090980529785, "train_loss_llm": 0.40659594535827637, "global_step": 98, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 0.880795955657959, "train_loss_dp": 0.8798403739929199, "train_loss_llm": 0.9555797576904297, "global_step": 99, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 0.976676344871521, "train_loss_dp": 0.9752496480941772, "train_loss_llm": 1.4266760349273682, "global_step": 100, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.062575101852417, "train_loss_dp": 1.0616531372070312, "train_loss_llm": 0.9219627380371094, "global_step": 101, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.163649559020996, "train_loss_dp": 1.1630744934082031, "train_loss_llm": 0.5750789642333984, "global_step": 102, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.1659646034240723, "train_loss_dp": 1.1644301414489746, "train_loss_llm": 1.5344963073730469, "global_step": 103, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.0758721828460693, "train_loss_dp": 1.07545804977417, "train_loss_llm": 0.4140939712524414, "global_step": 104, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.1677193641662598, "train_loss_dp": 1.166299819946289, "train_loss_llm": 1.4195492267608643, "global_step": 105, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 0.87457275390625, "train_loss_dp": 0.8743606805801392, "train_loss_llm": 0.2120532989501953, "global_step": 106, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 0.912935733795166, "train_loss_dp": 0.912591278553009, "train_loss_llm": 0.3444700241088867, "global_step": 107, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.1732856035232544, "train_loss_dp": 1.1709849834442139, "train_loss_llm": 2.3005619049072266, "global_step": 108, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 0.7645201683044434, "train_loss_dp": 0.7637201547622681, "train_loss_llm": 0.8000310659408569, "global_step": 109, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.176607608795166, "train_loss_dp": 1.1709827184677124, "train_loss_llm": 5.624931335449219, "global_step": 110, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.0543700456619263, "train_loss_dp": 1.0530896186828613, "train_loss_llm": 1.2804579734802246, "global_step": 111, "epoch": 0, "lr": 1.4000000000000001e-06}
+{"train_loss": 1.1628527641296387, "train_loss_dp": 1.1625275611877441, "train_loss_llm": 0.3252229690551758, "global_step": 112, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.3412309885025024, "train_loss_dp": 1.3376593589782715, "train_loss_llm": 3.5716705322265625, "global_step": 113, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 0.8261384963989258, "train_loss_dp": 0.8253236413002014, "train_loss_llm": 0.8148632049560547, "global_step": 114, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.010982871055603, "train_loss_dp": 1.010079264640808, "train_loss_llm": 0.903630256652832, "global_step": 115, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.054580807685852, "train_loss_dp": 1.0493314266204834, "train_loss_llm": 5.249382019042969, "global_step": 116, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.0486009120941162, "train_loss_dp": 1.0467684268951416, "train_loss_llm": 1.8324837684631348, "global_step": 117, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.0623599290847778, "train_loss_dp": 1.0612273216247559, "train_loss_llm": 1.132563591003418, "global_step": 118, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.055920124053955, "train_loss_dp": 1.0546772480010986, "train_loss_llm": 1.2429218292236328, "global_step": 119, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 0.9706803560256958, "train_loss_dp": 0.966740071773529, "train_loss_llm": 3.940277576446533, "global_step": 120, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 0.899577260017395, "train_loss_dp": 0.8988329172134399, "train_loss_llm": 0.744354248046875, "global_step": 121, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.0479618310928345, "train_loss_dp": 1.0471842288970947, "train_loss_llm": 0.7776050567626953, "global_step": 122, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 0.8802290558815002, "train_loss_dp": 0.8775439858436584, "train_loss_llm": 2.6850640773773193, "global_step": 123, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.3925269842147827, "train_loss_dp": 1.3899670839309692, "train_loss_llm": 2.559953212738037, "global_step": 124, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 0.7327772974967957, "train_loss_dp": 0.7316875457763672, "train_loss_llm": 1.0897562503814697, "global_step": 125, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 0.834916889667511, "train_loss_dp": 0.8332582116127014, "train_loss_llm": 1.6586933135986328, "global_step": 126, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 1.0088520050048828, "train_loss_dp": 1.0079145431518555, "train_loss_llm": 0.9374985694885254, "global_step": 127, "epoch": 0, "lr": 1.6000000000000001e-06}
+{"train_loss": 0.8526560664176941, "train_loss_dp": 0.8522714376449585, "train_loss_llm": 0.3846397399902344, "global_step": 128, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.0108627080917358, "train_loss_dp": 1.0083171129226685, "train_loss_llm": 2.54559326171875, "global_step": 129, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 0.7885764241218567, "train_loss_dp": 0.788051962852478, "train_loss_llm": 0.5244706869125366, "global_step": 130, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.273042917251587, "train_loss_dp": 1.2702627182006836, "train_loss_llm": 2.7802562713623047, "global_step": 131, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.1908512115478516, "train_loss_dp": 1.1891257762908936, "train_loss_llm": 1.7254295349121094, "global_step": 132, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 0.9060755968093872, "train_loss_dp": 0.9055706858634949, "train_loss_llm": 0.5049176216125488, "global_step": 133, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 0.9288762807846069, "train_loss_dp": 0.9278987646102905, "train_loss_llm": 0.9775400161743164, "global_step": 134, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.0303034782409668, "train_loss_dp": 1.0282714366912842, "train_loss_llm": 2.0320777893066406, "global_step": 135, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.07591712474823, "train_loss_dp": 1.073346495628357, "train_loss_llm": 2.570608139038086, "global_step": 136, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 0.7770967483520508, "train_loss_dp": 0.7760414481163025, "train_loss_llm": 1.055305004119873, "global_step": 137, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 0.9279473423957825, "train_loss_dp": 0.9250935912132263, "train_loss_llm": 2.853731870651245, "global_step": 138, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.0239455699920654, "train_loss_dp": 1.0234155654907227, "train_loss_llm": 0.5299513339996338, "global_step": 139, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 0.6870459914207458, "train_loss_dp": 0.685171365737915, "train_loss_llm": 1.874603271484375, "global_step": 140, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.0657446384429932, "train_loss_dp": 1.0648167133331299, "train_loss_llm": 0.9278755187988281, "global_step": 141, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.12233567237854, "train_loss_dp": 1.1173756122589111, "train_loss_llm": 4.960109710693359, "global_step": 142, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 1.3854516744613647, "train_loss_dp": 1.384528398513794, "train_loss_llm": 0.9232254028320312, "global_step": 143, "epoch": 0, "lr": 1.8e-06}
+{"train_loss": 0.9985589385032654, "train_loss_dp": 0.9981729388237, "train_loss_llm": 0.3860071897506714, "global_step": 144, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.1671257019042969, "train_loss_dp": 1.1661062240600586, "train_loss_llm": 1.0194945335388184, "global_step": 145, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 0.9584890604019165, "train_loss_dp": 0.9580555558204651, "train_loss_llm": 0.4334831237792969, "global_step": 146, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.2573744058609009, "train_loss_dp": 1.2566215991973877, "train_loss_llm": 0.7527618408203125, "global_step": 147, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 0.9796014428138733, "train_loss_dp": 0.9774230122566223, "train_loss_llm": 2.1784095764160156, "global_step": 148, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.0690115690231323, "train_loss_dp": 1.0686674118041992, "train_loss_llm": 0.34418439865112305, "global_step": 149, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.3584924936294556, "train_loss_dp": 1.3569716215133667, "train_loss_llm": 1.5209159851074219, "global_step": 150, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.319930911064148, "train_loss_dp": 1.3195158243179321, "train_loss_llm": 0.4150766134262085, "global_step": 151, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.373637318611145, "train_loss_dp": 1.3714826107025146, "train_loss_llm": 2.1546716690063477, "global_step": 152, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 0.7981638312339783, "train_loss_dp": 0.7972655296325684, "train_loss_llm": 0.8983020782470703, "global_step": 153, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 0.6259526610374451, "train_loss_dp": 0.6259526610374451, "train_loss_llm": 0.0, "global_step": 154, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 0.8991695642471313, "train_loss_dp": 0.8981265425682068, "train_loss_llm": 1.0430073738098145, "global_step": 155, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 0.8341158032417297, "train_loss_dp": 0.8331246376037598, "train_loss_llm": 0.9911479949951172, "global_step": 156, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.3519165515899658, "train_loss_dp": 1.349484920501709, "train_loss_llm": 2.4316177368164062, "global_step": 157, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.174717903137207, "train_loss_dp": 1.1742562055587769, "train_loss_llm": 0.46164369583129883, "global_step": 158, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 1.1371837854385376, "train_loss_dp": 1.1310091018676758, "train_loss_llm": 6.17466926574707, "global_step": 159, "epoch": 0, "lr": 2.0000000000000003e-06}
+{"train_loss": 0.9435285329818726, "train_loss_dp": 0.9389777183532715, "train_loss_llm": 4.5507965087890625, "global_step": 160, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 0.9228729605674744, "train_loss_dp": 0.9212661981582642, "train_loss_llm": 1.6067745685577393, "global_step": 161, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.0533033609390259, "train_loss_dp": 1.0522117614746094, "train_loss_llm": 1.0916099548339844, "global_step": 162, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 0.9917002320289612, "train_loss_dp": 0.990831732749939, "train_loss_llm": 0.868518590927124, "global_step": 163, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.2273536920547485, "train_loss_dp": 1.2264316082000732, "train_loss_llm": 0.9221370220184326, "global_step": 164, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 0.8994293808937073, "train_loss_dp": 0.8975485563278198, "train_loss_llm": 1.8808438777923584, "global_step": 165, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.0688377618789673, "train_loss_dp": 1.067429780960083, "train_loss_llm": 1.408036231994629, "global_step": 166, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.0449057817459106, "train_loss_dp": 1.0439999103546143, "train_loss_llm": 0.9059271812438965, "global_step": 167, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 0.9799196124076843, "train_loss_dp": 0.9787405133247375, "train_loss_llm": 1.1790766716003418, "global_step": 168, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 0.9629499316215515, "train_loss_dp": 0.9624240398406982, "train_loss_llm": 0.5258660316467285, "global_step": 169, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.0641788244247437, "train_loss_dp": 1.0627567768096924, "train_loss_llm": 1.4221014976501465, "global_step": 170, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 0.668416440486908, "train_loss_dp": 0.6656877994537354, "train_loss_llm": 2.728656768798828, "global_step": 171, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 0.9890009760856628, "train_loss_dp": 0.986947774887085, "train_loss_llm": 2.053201675415039, "global_step": 172, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.248491883277893, "train_loss_dp": 1.2451238632202148, "train_loss_llm": 3.3679628372192383, "global_step": 173, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.0831762552261353, "train_loss_dp": 1.0828850269317627, "train_loss_llm": 0.2912263870239258, "global_step": 174, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 1.2579056024551392, "train_loss_dp": 1.256361961364746, "train_loss_llm": 1.5436620712280273, "global_step": 175, "epoch": 0, "lr": 2.2e-06}
+{"train_loss": 0.8215212225914001, "train_loss_dp": 0.820573091506958, "train_loss_llm": 0.9481277465820312, "global_step": 176, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.147844672203064, "train_loss_dp": 1.1466038227081299, "train_loss_llm": 1.2408628463745117, "global_step": 177, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 0.672270655632019, "train_loss_dp": 0.6717621088027954, "train_loss_llm": 0.508547306060791, "global_step": 178, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.225612759590149, "train_loss_dp": 1.2236188650131226, "train_loss_llm": 1.993865966796875, "global_step": 179, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.0962371826171875, "train_loss_dp": 1.0915204286575317, "train_loss_llm": 4.716758728027344, "global_step": 180, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 0.9209144711494446, "train_loss_dp": 0.9196658134460449, "train_loss_llm": 1.2486529350280762, "global_step": 181, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 0.8987254500389099, "train_loss_dp": 0.8983194828033447, "train_loss_llm": 0.40599536895751953, "global_step": 182, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.3857333660125732, "train_loss_dp": 1.3848227262496948, "train_loss_llm": 0.9106874465942383, "global_step": 183, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.0813459157943726, "train_loss_dp": 1.0796198844909668, "train_loss_llm": 1.7260627746582031, "global_step": 184, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.2093092203140259, "train_loss_dp": 1.2086422443389893, "train_loss_llm": 0.6669338941574097, "global_step": 185, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.0938993692398071, "train_loss_dp": 1.0907189846038818, "train_loss_llm": 3.1804208755493164, "global_step": 186, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 0.815351128578186, "train_loss_dp": 0.8142208456993103, "train_loss_llm": 1.1302626132965088, "global_step": 187, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.0570591688156128, "train_loss_dp": 1.0565214157104492, "train_loss_llm": 0.5377693176269531, "global_step": 188, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 0.8795514702796936, "train_loss_dp": 0.8777532577514648, "train_loss_llm": 1.798233985900879, "global_step": 189, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.3283787965774536, "train_loss_dp": 1.3247414827346802, "train_loss_llm": 3.637310743331909, "global_step": 190, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.0419065952301025, "train_loss_dp": 1.0408300161361694, "train_loss_llm": 1.0766291618347168, "global_step": 191, "epoch": 0, "lr": 2.4000000000000003e-06}
+{"train_loss": 1.2874374389648438, "train_loss_dp": 1.2858664989471436, "train_loss_llm": 1.5709571838378906, "global_step": 192, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.201712965965271, "train_loss_dp": 1.200538158416748, "train_loss_llm": 1.1747956275939941, "global_step": 193, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.1180351972579956, "train_loss_dp": 1.1164162158966064, "train_loss_llm": 1.6189470291137695, "global_step": 194, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.253013253211975, "train_loss_dp": 1.2514146566390991, "train_loss_llm": 1.5985565185546875, "global_step": 195, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.0498392581939697, "train_loss_dp": 1.0492035150527954, "train_loss_llm": 0.6357004642486572, "global_step": 196, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 0.9419779181480408, "train_loss_dp": 0.9411032199859619, "train_loss_llm": 0.874704122543335, "global_step": 197, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.384355068206787, "train_loss_dp": 1.383790373802185, "train_loss_llm": 0.564707338809967, "global_step": 198, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.0896580219268799, "train_loss_dp": 1.0865449905395508, "train_loss_llm": 3.1130428314208984, "global_step": 199, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 0.7918712496757507, "train_loss_dp": 0.7875099182128906, "train_loss_llm": 4.3613433837890625, "global_step": 200, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 0.6694862246513367, "train_loss_dp": 0.6685686707496643, "train_loss_llm": 0.9175718426704407, "global_step": 201, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 0.7734841704368591, "train_loss_dp": 0.7727508544921875, "train_loss_llm": 0.7333073019981384, "global_step": 202, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.0277655124664307, "train_loss_dp": 1.026642084121704, "train_loss_llm": 1.1233766078948975, "global_step": 203, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 0.9652549028396606, "train_loss_dp": 0.9644122123718262, "train_loss_llm": 0.8427028656005859, "global_step": 204, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 0.8998804688453674, "train_loss_dp": 0.8987721800804138, "train_loss_llm": 1.1082992553710938, "global_step": 205, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.0590038299560547, "train_loss_dp": 1.0581740140914917, "train_loss_llm": 0.829864501953125, "global_step": 206, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.1207412481307983, "train_loss_dp": 1.120018482208252, "train_loss_llm": 0.7227258682250977, "global_step": 207, "epoch": 0, "lr": 2.6e-06}
+{"train_loss": 1.1778478622436523, "train_loss_dp": 1.1764674186706543, "train_loss_llm": 1.3804551362991333, "global_step": 208, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.0366995334625244, "train_loss_dp": 1.0359495878219604, "train_loss_llm": 0.7498936653137207, "global_step": 209, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.257690191268921, "train_loss_dp": 1.2571049928665161, "train_loss_llm": 0.5852546691894531, "global_step": 210, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.2400319576263428, "train_loss_dp": 1.2349765300750732, "train_loss_llm": 5.055484771728516, "global_step": 211, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.9779812097549438, "train_loss_dp": 0.9769306182861328, "train_loss_llm": 1.050614356994629, "global_step": 212, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.026679277420044, "train_loss_dp": 1.025492787361145, "train_loss_llm": 1.1865310668945312, "global_step": 213, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.070602536201477, "train_loss_dp": 1.0699427127838135, "train_loss_llm": 0.6598005294799805, "global_step": 214, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.958576500415802, "train_loss_dp": 0.9577209949493408, "train_loss_llm": 0.8555289506912231, "global_step": 215, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.8888511657714844, "train_loss_dp": 0.8880195021629333, "train_loss_llm": 0.8316879272460938, "global_step": 216, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.9753172397613525, "train_loss_dp": 0.9743190407752991, "train_loss_llm": 0.9982223510742188, "global_step": 217, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.060932993888855, "train_loss_dp": 1.059741497039795, "train_loss_llm": 1.1914410591125488, "global_step": 218, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.2638167142868042, "train_loss_dp": 1.260404109954834, "train_loss_llm": 3.412565231323242, "global_step": 219, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.9427530169487, "train_loss_dp": 0.9422386884689331, "train_loss_llm": 0.5143337249755859, "global_step": 220, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.0893369913101196, "train_loss_dp": 1.0871870517730713, "train_loss_llm": 2.1499853134155273, "global_step": 221, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.0557167530059814, "train_loss_dp": 1.0530527830123901, "train_loss_llm": 2.6639106273651123, "global_step": 222, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 1.0937981605529785, "train_loss_dp": 1.0928211212158203, "train_loss_llm": 0.9770201444625854, "global_step": 223, "epoch": 0, "lr": 2.8000000000000003e-06}
+{"train_loss": 0.8541648387908936, "train_loss_dp": 0.8535228371620178, "train_loss_llm": 0.6420044898986816, "global_step": 224, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.206831693649292, "train_loss_dp": 1.2036032676696777, "train_loss_llm": 3.2283689975738525, "global_step": 225, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.1192035675048828, "train_loss_dp": 1.1170153617858887, "train_loss_llm": 2.1882400512695312, "global_step": 226, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.4599449634552002, "train_loss_dp": 1.45904541015625, "train_loss_llm": 0.8995132446289062, "global_step": 227, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.0684638023376465, "train_loss_dp": 1.065883994102478, "train_loss_llm": 2.5798492431640625, "global_step": 228, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.1733022928237915, "train_loss_dp": 1.1723358631134033, "train_loss_llm": 0.9664278030395508, "global_step": 229, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.8719821572303772, "train_loss_dp": 0.8709094524383545, "train_loss_llm": 1.072675108909607, "global_step": 230, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.083701252937317, "train_loss_dp": 1.0802502632141113, "train_loss_llm": 3.451019287109375, "global_step": 231, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.8477475643157959, "train_loss_dp": 0.847568690776825, "train_loss_llm": 0.17889785766601562, "global_step": 232, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.8178531527519226, "train_loss_dp": 0.817230761051178, "train_loss_llm": 0.6224136352539062, "global_step": 233, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.3234915733337402, "train_loss_dp": 1.3229807615280151, "train_loss_llm": 0.5107967853546143, "global_step": 234, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.9152907729148865, "train_loss_dp": 0.9146604537963867, "train_loss_llm": 0.6303447484970093, "global_step": 235, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.4234950542449951, "train_loss_dp": 1.396744728088379, "train_loss_llm": 26.750328063964844, "global_step": 236, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.1292498111724854, "train_loss_dp": 1.1266402006149292, "train_loss_llm": 2.6095657348632812, "global_step": 237, "epoch": 0, "lr": 3e-06}
+{"train_loss": 0.9250983595848083, "train_loss_dp": 0.9231075644493103, "train_loss_llm": 1.9908084869384766, "global_step": 238, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.3176473379135132, "train_loss_dp": 1.3054389953613281, "train_loss_llm": 12.208351135253906, "global_step": 239, "epoch": 0, "lr": 3e-06}
+{"train_loss": 1.1565420627593994, "train_loss_dp": 1.1558828353881836, "train_loss_llm": 0.6591784954071045, "global_step": 240, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.8393681049346924, "train_loss_dp": 0.8389867544174194, "train_loss_llm": 0.3813209533691406, "global_step": 241, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.0412899255752563, "train_loss_dp": 1.0390691757202148, "train_loss_llm": 2.220743179321289, "global_step": 242, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.8673443794250488, "train_loss_dp": 0.8664416670799255, "train_loss_llm": 0.902684211730957, "global_step": 243, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.7161498069763184, "train_loss_dp": 0.7161498069763184, "train_loss_llm": 0.0, "global_step": 244, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.0614242553710938, "train_loss_dp": 1.0592397451400757, "train_loss_llm": 2.1844592094421387, "global_step": 245, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.7298458814620972, "train_loss_dp": 0.7288863658905029, "train_loss_llm": 0.9595050811767578, "global_step": 246, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.9787002801895142, "train_loss_dp": 0.977652907371521, "train_loss_llm": 1.0473737716674805, "global_step": 247, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.9223641157150269, "train_loss_dp": 0.9200658798217773, "train_loss_llm": 2.2982349395751953, "global_step": 248, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.8517196178436279, "train_loss_dp": 0.8488768339157104, "train_loss_llm": 2.842813491821289, "global_step": 249, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.7904592156410217, "train_loss_dp": 0.7890785336494446, "train_loss_llm": 1.3807096481323242, "global_step": 250, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 1.022873044013977, "train_loss_dp": 1.0225629806518555, "train_loss_llm": 0.31003570556640625, "global_step": 251, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.8171392679214478, "train_loss_dp": 0.8164294362068176, "train_loss_llm": 0.7098064422607422, "global_step": 252, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.9496151804924011, "train_loss_dp": 0.9483891725540161, "train_loss_llm": 1.2259790897369385, "global_step": 253, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.9365168809890747, "train_loss_dp": 0.9358597993850708, "train_loss_llm": 0.6570758819580078, "global_step": 254, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.7296798229217529, "train_loss_dp": 0.7292571663856506, "train_loss_llm": 0.4226555824279785, "global_step": 255, "epoch": 0, "lr": 3.2000000000000003e-06}
+{"train_loss": 0.8417019248008728, "train_loss_dp": 0.8404977321624756, "train_loss_llm": 1.2041740417480469, "global_step": 256, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.0454022884368896, "train_loss_dp": 1.0439611673355103, "train_loss_llm": 1.4411430358886719, "global_step": 257, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.1510649919509888, "train_loss_dp": 1.1502982378005981, "train_loss_llm": 0.7667382955551147, "global_step": 258, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.035636067390442, "train_loss_dp": 1.0333507061004639, "train_loss_llm": 2.2853426933288574, "global_step": 259, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.0258796215057373, "train_loss_dp": 1.0221712589263916, "train_loss_llm": 3.7083663940429688, "global_step": 260, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.2506957054138184, "train_loss_dp": 1.248490810394287, "train_loss_llm": 2.2049503326416016, "global_step": 261, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.263920783996582, "train_loss_dp": 1.2621973752975464, "train_loss_llm": 1.7233924865722656, "global_step": 262, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.315664291381836, "train_loss_dp": 1.3132370710372925, "train_loss_llm": 2.427274227142334, "global_step": 263, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 0.8286254405975342, "train_loss_dp": 0.827805757522583, "train_loss_llm": 0.8196887969970703, "global_step": 264, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.3872648477554321, "train_loss_dp": 1.3866569995880127, "train_loss_llm": 0.6078777313232422, "global_step": 265, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.1013479232788086, "train_loss_dp": 1.0999729633331299, "train_loss_llm": 1.3749570846557617, "global_step": 266, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.1393728256225586, "train_loss_dp": 1.1385104656219482, "train_loss_llm": 0.8623199462890625, "global_step": 267, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 0.9952676296234131, "train_loss_dp": 0.9947441816329956, "train_loss_llm": 0.5234229564666748, "global_step": 268, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.0814908742904663, "train_loss_dp": 1.0808988809585571, "train_loss_llm": 0.5920498371124268, "global_step": 269, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.1242868900299072, "train_loss_dp": 1.1228426694869995, "train_loss_llm": 1.4442572593688965, "global_step": 270, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 0.9303393959999084, "train_loss_dp": 0.9291984438896179, "train_loss_llm": 1.1409424543380737, "global_step": 271, "epoch": 0, "lr": 3.4000000000000005e-06}
+{"train_loss": 1.053684115409851, "train_loss_dp": 1.0524609088897705, "train_loss_llm": 1.2232160568237305, "global_step": 272, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.8706421256065369, "train_loss_dp": 0.8694562315940857, "train_loss_llm": 1.1858901977539062, "global_step": 273, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.981246292591095, "train_loss_dp": 0.9806689023971558, "train_loss_llm": 0.577362060546875, "global_step": 274, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.2208939790725708, "train_loss_dp": 1.2203340530395508, "train_loss_llm": 0.5598692893981934, "global_step": 275, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.31655752658844, "train_loss_dp": 1.316185474395752, "train_loss_llm": 0.3720601201057434, "global_step": 276, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.9832253456115723, "train_loss_dp": 0.981768786907196, "train_loss_llm": 1.456533432006836, "global_step": 277, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.7552657723426819, "train_loss_dp": 0.7519952654838562, "train_loss_llm": 3.270486831665039, "global_step": 278, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.1233352422714233, "train_loss_dp": 1.1223489046096802, "train_loss_llm": 0.9863491058349609, "global_step": 279, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.8886633515357971, "train_loss_dp": 0.8878415822982788, "train_loss_llm": 0.821770191192627, "global_step": 280, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.9254290461540222, "train_loss_dp": 0.9251669645309448, "train_loss_llm": 0.2620983123779297, "global_step": 281, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.1799665689468384, "train_loss_dp": 1.1799665689468384, "train_loss_llm": 0.0, "global_step": 282, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.9371044635772705, "train_loss_dp": 0.9369461536407471, "train_loss_llm": 0.15830063819885254, "global_step": 283, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.7163202166557312, "train_loss_dp": 0.7144418954849243, "train_loss_llm": 1.8782930374145508, "global_step": 284, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.7892820835113525, "train_loss_dp": 0.7881895303726196, "train_loss_llm": 1.0925536155700684, "global_step": 285, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.9342642426490784, "train_loss_dp": 0.9331384897232056, "train_loss_llm": 1.1257610321044922, "global_step": 286, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 0.7539575695991516, "train_loss_dp": 0.7533442974090576, "train_loss_llm": 0.6133006811141968, "global_step": 287, "epoch": 0, "lr": 3.6e-06}
+{"train_loss": 1.2160762548446655, "train_loss_dp": 1.2145178318023682, "train_loss_llm": 1.558393955230713, "global_step": 288, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.0810425281524658, "train_loss_dp": 1.0801011323928833, "train_loss_llm": 0.9414329528808594, "global_step": 289, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.2203058004379272, "train_loss_dp": 1.2177138328552246, "train_loss_llm": 2.5920188426971436, "global_step": 290, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.0211195945739746, "train_loss_dp": 1.0205960273742676, "train_loss_llm": 0.5235177278518677, "global_step": 291, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.0563230514526367, "train_loss_dp": 1.0549695491790771, "train_loss_llm": 1.3535313606262207, "global_step": 292, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.6651569604873657, "train_loss_dp": 0.6641620993614197, "train_loss_llm": 0.9948391914367676, "global_step": 293, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.7676696181297302, "train_loss_dp": 0.7625374794006348, "train_loss_llm": 5.132167816162109, "global_step": 294, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.8271551728248596, "train_loss_dp": 0.8251689672470093, "train_loss_llm": 1.986194133758545, "global_step": 295, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.9217114448547363, "train_loss_dp": 0.9022454023361206, "train_loss_llm": 19.466018676757812, "global_step": 296, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.1559274196624756, "train_loss_dp": 1.1552963256835938, "train_loss_llm": 0.6311402320861816, "global_step": 297, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.2436766624450684, "train_loss_dp": 1.2431010007858276, "train_loss_llm": 0.5756945610046387, "global_step": 298, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.6809545159339905, "train_loss_dp": 0.6802892088890076, "train_loss_llm": 0.6653022766113281, "global_step": 299, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.7953881025314331, "train_loss_dp": 0.794620156288147, "train_loss_llm": 0.7679550647735596, "global_step": 300, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.9237709641456604, "train_loss_dp": 0.921987771987915, "train_loss_llm": 1.7832164764404297, "global_step": 301, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.9575345516204834, "train_loss_dp": 0.956720232963562, "train_loss_llm": 0.8143423795700073, "global_step": 302, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 1.0717120170593262, "train_loss_dp": 1.0692625045776367, "train_loss_llm": 2.449453353881836, "global_step": 303, "epoch": 0, "lr": 3.8e-06}
+{"train_loss": 0.6543203592300415, "train_loss_dp": 0.6536703109741211, "train_loss_llm": 0.6500511169433594, "global_step": 304, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.0165913105010986, "train_loss_dp": 1.0153517723083496, "train_loss_llm": 1.2394866943359375, "global_step": 305, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.785950779914856, "train_loss_dp": 0.7853435277938843, "train_loss_llm": 0.607224702835083, "global_step": 306, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.1852357387542725, "train_loss_dp": 1.184282660484314, "train_loss_llm": 0.9530715942382812, "global_step": 307, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.3600327968597412, "train_loss_dp": 1.3591878414154053, "train_loss_llm": 0.8449587821960449, "global_step": 308, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.8652999401092529, "train_loss_dp": 0.8323477506637573, "train_loss_llm": 32.952171325683594, "global_step": 309, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.0732994079589844, "train_loss_dp": 1.0726039409637451, "train_loss_llm": 0.69547438621521, "global_step": 310, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.7763187885284424, "train_loss_dp": 0.7752112150192261, "train_loss_llm": 1.1075663566589355, "global_step": 311, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.804984986782074, "train_loss_dp": 0.8042407035827637, "train_loss_llm": 0.7442779541015625, "global_step": 312, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.03466796875, "train_loss_dp": 1.0335252285003662, "train_loss_llm": 1.14276123046875, "global_step": 313, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.9160119891166687, "train_loss_dp": 0.9146207571029663, "train_loss_llm": 1.3912560939788818, "global_step": 314, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.250584602355957, "train_loss_dp": 1.249985933303833, "train_loss_llm": 0.5987164974212646, "global_step": 315, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.7121689319610596, "train_loss_dp": 0.711330771446228, "train_loss_llm": 0.838179349899292, "global_step": 316, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.9958481192588806, "train_loss_dp": 0.9951506853103638, "train_loss_llm": 0.6974499225616455, "global_step": 317, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.7193962335586548, "train_loss_dp": 0.7181667685508728, "train_loss_llm": 1.2294502258300781, "global_step": 318, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 0.820712685585022, "train_loss_dp": 0.8187174201011658, "train_loss_llm": 1.9952507019042969, "global_step": 319, "epoch": 0, "lr": 4.000000000000001e-06}
+{"train_loss": 1.0842673778533936, "train_loss_dp": 1.0823585987091064, "train_loss_llm": 1.9088265895843506, "global_step": 320, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.0535521507263184, "train_loss_dp": 1.0532820224761963, "train_loss_llm": 0.27007293701171875, "global_step": 321, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.7156171798706055, "train_loss_dp": 0.7132514119148254, "train_loss_llm": 2.3657703399658203, "global_step": 322, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.3940080404281616, "train_loss_dp": 1.3935505151748657, "train_loss_llm": 0.45752644538879395, "global_step": 323, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.6467493176460266, "train_loss_dp": 0.6460374593734741, "train_loss_llm": 0.7118768692016602, "global_step": 324, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.0929086208343506, "train_loss_dp": 1.092090368270874, "train_loss_llm": 0.818206787109375, "global_step": 325, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.9168623089790344, "train_loss_dp": 0.9167463779449463, "train_loss_llm": 0.11590766906738281, "global_step": 326, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.050483226776123, "train_loss_dp": 1.0486924648284912, "train_loss_llm": 1.7907218933105469, "global_step": 327, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.895656168460846, "train_loss_dp": 0.8935145139694214, "train_loss_llm": 2.1416430473327637, "global_step": 328, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.9489437937736511, "train_loss_dp": 0.9478486776351929, "train_loss_llm": 1.0950896739959717, "global_step": 329, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.7653458118438721, "train_loss_dp": 0.7636327147483826, "train_loss_llm": 1.7131099700927734, "global_step": 330, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.0067107677459717, "train_loss_dp": 1.005708932876587, "train_loss_llm": 1.0018339157104492, "global_step": 331, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.8278260827064514, "train_loss_dp": 0.8270441889762878, "train_loss_llm": 0.781912088394165, "global_step": 332, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.1179126501083374, "train_loss_dp": 1.1171658039093018, "train_loss_llm": 0.746841311454773, "global_step": 333, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.9853150248527527, "train_loss_dp": 0.9846751093864441, "train_loss_llm": 0.6399421691894531, "global_step": 334, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 1.1449363231658936, "train_loss_dp": 1.1282262802124023, "train_loss_llm": 16.709983825683594, "global_step": 335, "epoch": 0, "lr": 4.2000000000000004e-06}
+{"train_loss": 0.907992959022522, "train_loss_dp": 0.9071063995361328, "train_loss_llm": 0.886573076248169, "global_step": 336, "epoch": 0, "lr": 4.4e-06}
+{"train_loss": 0.7433826327323914, "train_loss_dp": 0.7433826327323914, "train_loss_llm": 0.0, "global_step": 337, "epoch": 0, "lr": 4.4e-06}

2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/train.log

@@ -0,0 +1,11 @@
+[2026-04-29 08:36:57,609][numexpr.utils][INFO] - Note: NumExpr detected 36 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 16.
+[2026-04-29 08:36:57,609][numexpr.utils][INFO] - NumExpr defaulting to 16 threads.
+[2026-04-29 08:37:00,053][datasets][INFO] - PyTorch version 2.2.2 available.
+[2026-04-29 08:37:00,054][datasets][INFO] - TensorFlow version 2.15.1 available.
+[2026-04-29 08:37:00,054][datasets][INFO] - JAX version 0.4.30 available.
+[2026-04-29 08:37:01,783][llmbc.model.diffusion.conditional_unet1d][INFO] - number of parameters: 6.501479e+07
+[2026-04-29 08:37:18,087][root][INFO] - running build_ext
+[2026-04-29 08:37:18,090][root][INFO] - building 'mujoco_py.cymj' extension
+[2026-04-29 08:37:18,090][root][INFO] - gcc -pthread -B /venv/llm-bc/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /venv/llm-bc/include -I/venv/llm-bc/include -fPIC -O2 -isystem /venv/llm-bc/include -fPIC -I/venv/llm-bc/lib/python3.9/site-packages/mujoco_py -I/root/.mujoco/mujoco210/include -I/venv/llm-bc/lib/python3.9/site-packages/numpy/core/include -I/venv/llm-bc/include/python3.9 -c /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/cymj.c -o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/cymj.o -fopenmp -w
+[2026-04-29 08:37:45,478][root][INFO] - gcc -pthread -B /venv/llm-bc/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /venv/llm-bc/include -I/venv/llm-bc/include -fPIC -O2 -isystem /venv/llm-bc/include -fPIC -I/venv/llm-bc/lib/python3.9/site-packages/mujoco_py -I/root/.mujoco/mujoco210/include -I/venv/llm-bc/lib/python3.9/site-packages/numpy/core/include -I/venv/llm-bc/include/python3.9 -c /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/gl/osmesashim.c -o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/gl/osmesashim.o -fopenmp -w
+[2026-04-29 08:37:45,623][root][INFO] - gcc -pthread -B /venv/llm-bc/compiler_compat -shared -Wl,-rpath,/venv/llm-bc/lib -Wl,-rpath-link,/venv/llm-bc/lib -L/venv/llm-bc/lib -L/venv/llm-bc/lib -Wl,-rpath,/venv/llm-bc/lib -Wl,-rpath-link,/venv/llm-bc/lib -L/venv/llm-bc/lib /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/cymj.o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/temp.linux-x86_64-cpython-39/venv/llm-bc/lib/python3.9/site-packages/mujoco_py/gl/osmesashim.o -L/root/.mujoco/mujoco210/bin -Wl,--enable-new-dtags,-rpath,/root/.mujoco/mujoco210/bin -lmujoco210 -lglewosmesa -lOSMesa -lGL -o /venv/llm-bc/lib/python3.9/site-packages/mujoco_py/generated/_pyxbld_2.1.2.14_39_linuxcpuextensionbuilder/lib.linux-x86_64-cpython-39/mujoco_py/cymj.cpython-39-x86_64-linux-gnu.so -fopenmp

2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083749-n45frqzz/files/output.log

@@ -0,0 +1,72 @@
+Traceback (most recent call last):                                                                                                                                                    
+  File "/workspace/LLM-BC/./train.py", line 35, in <module>
+    main()
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/main.py", line 90, in decorated_main
+    _run_hydra(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 389, in _run_hydra
+    _run_app(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 452, in _run_app
+    run_and_report(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 213, in run_and_report
+    return func()
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 453, in <lambda>
+    lambda: hydra.run(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/hydra.py", line 119, in run
+    ret = run_job(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/core/utils.py", line 186, in run_job
+    ret.return_value = task_function(task_cfg)
+  File "/workspace/LLM-BC/./train.py", line 32, in main
+    workspace.run()
+  File "/workspace/LLM-BC/llmbc/workspace/train_llm_diffusion_unet_lowdim_workspace.py", line 252, in run
+    runner_log = env_runner.run(policy)
+  File "/workspace/LLM-BC/llmbc/env_runner/metaworld_lowdim_runner.py", line 113, in run
+    self.env = AsyncVectorEnv(self.env_fns)
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 168, in __init__
+    self._check_observation_spaces()
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 378, in _check_observation_spaces
+    same_spaces, successes = zip(*[pipe.recv() for pipe in self.parent_pipes])
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 378, in <listcomp>
+    same_spaces, successes = zip(*[pipe.recv() for pipe in self.parent_pipes])
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 250, in recv
+    buf = self._recv_bytes()
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 414, in _recv_bytes
+    buf = self._recv(4)
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 379, in _recv
+    chunk = read(handle, remaining)
+KeyboardInterrupt
+Traceback (most recent call last):
+  File "/workspace/LLM-BC/./train.py", line 35, in <module>
+    main()
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/main.py", line 90, in decorated_main
+    _run_hydra(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 389, in _run_hydra
+    _run_app(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 452, in _run_app
+    run_and_report(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 213, in run_and_report
+    return func()
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/utils.py", line 453, in <lambda>
+    lambda: hydra.run(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/_internal/hydra.py", line 119, in run
+    ret = run_job(
+  File "/venv/llm-bc/lib/python3.9/site-packages/hydra/core/utils.py", line 186, in run_job
+    ret.return_value = task_function(task_cfg)
+  File "/workspace/LLM-BC/./train.py", line 32, in main
+    workspace.run()
+  File "/workspace/LLM-BC/llmbc/workspace/train_llm_diffusion_unet_lowdim_workspace.py", line 252, in run
+    runner_log = env_runner.run(policy)
+  File "/workspace/LLM-BC/llmbc/env_runner/metaworld_lowdim_runner.py", line 113, in run
+    self.env = AsyncVectorEnv(self.env_fns)
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 168, in __init__
+    self._check_observation_spaces()
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 378, in _check_observation_spaces
+    same_spaces, successes = zip(*[pipe.recv() for pipe in self.parent_pipes])
+  File "/workspace/LLM-BC/llmbc/gym_util/async_vector_env.py", line 378, in <listcomp>
+    same_spaces, successes = zip(*[pipe.recv() for pipe in self.parent_pipes])
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 250, in recv
+    buf = self._recv_bytes()
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 414, in _recv_bytes
+    buf = self._recv(4)
+  File "/venv/llm-bc/lib/python3.9/multiprocessing/connection.py", line 379, in _recv
+    chunk = read(handle, remaining)
+KeyboardInterrupt

2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083749-n45frqzz/files/requirements.txt

@@ -0,0 +1,299 @@
+llmbc==0.0.0
+aiohappyeyeballs==2.6.1
+antlr4-python3-runtime==4.9.3
+asciitree==0.3.3
+async-timeout==5.0.1
+attrs==25.4.0
+certifi==2025.10.5
+charset-normalizer==3.4.4
+colorama==0.4.6
+dill==0.3.8
+docstring-parser==0.16
+eval_type_backport==0.2.2
+fasteners==0.20
+filelock==3.17.0
+frozenlist==1.8.0
+fsspec==2024.3.1
+gmpy2==2.2.1
+hf-xet==1.1.8
+idna==3.11
+llvmlite==0.39.1
+MarkupSafe==3.0.2
+mdurl==0.1.2
+mkl-service==2.4.0
+mpmath==1.3.0
+msgpack==1.1.1
+networkx==3.2.1
+packaging==25.0
+propcache==0.3.1
+psutil==7.0.0
+pycparser==2.23
+Pygments==2.19.1
+PyOpenGL==3.1.10
+PySocks==1.7.1
+tzdata==2025.3
+xxhash==3.5.0
+pytz==2025.2
+PyYAML==6.0.2
+regex==2025.9.1
+safetensors==0.5.3
+setuptools==80.9.0
+shtab==1.7.2
+six==1.17.0
+tqdm==4.67.1
+tensorboard==2.15.2
+uvloop==0.22.1
+wheel==0.45.1
+zipp==3.21.0
+tensorflow-estimator==2.15.0
+tensorboard-data-server==0.7.2
+google-auth-oauthlib==1.3.1
+tensorflow==2.15.1
+Jinja2==3.1.6
+markdown-it-py==2.2.0
+multiprocess==0.70.15
+numpy==1.23.5
+pillow==11.3.0
+pip==23.3.2
+python-dateutil==2.9.0.post0
+aiosignal==1.4.0
+brotlicffi==1.0.9.2
+hydra-core==1.2.0
+multidict==6.7.0
+pycares==4.10.0
+rich==14.2.0
+sympy==1.14.0
+typeguard==4.4.4
+aiodns==3.5.0
+urllib3==2.5.0
+yarl==1.18.0
+aiohttp==3.12.15
+pyarrow==21.0.0
+pyarrow-hotfix==0.7
+tokenizers==0.21.0
+av==10.0.0
+einops==0.4.1
+Bottleneck==1.4.2
+mkl_fft==1.3.11
+mkl_random==1.2.8
+numba==0.56.4
+numcodecs==0.12.1
+numexpr==2.10.1
+pandas==2.3.3
+zarr==2.12.0
+datasets==2.19.0
+transformers==4.47.1
+accelerate==1.0.1
+peft==0.14.0
+trl==0.11.4
+torch==2.2.2
+tabulate==0.9.0
+sentencepiece==0.2.0
+pyperclip==1.11.0
+PyOpenGL-accelerate==3.1.10
+pydub==0.25.1
+py-cpuinfo==9.0.0
+pure_eval==0.2.3
+ptyprocess==0.7.0
+nvidia-ml-py==13.595.45
+hjson==3.1.0
+gym-notices==0.1.0
+glfw==2.0.0
+Farama-Notifications==0.0.6
+websockets==15.0.1
+wcwidth==0.6.0
+uc-micro-py==1.0.3
+typing-inspection==0.4.2
+triton==2.2.0
+trimesh==4.12.1
+transforms3d==0.4.2
+traitlets==5.14.3
+threadpoolctl==3.6.0
+tenacity==9.1.2
+sniffio==1.3.1
+smmap==5.0.3
+setproctitle==1.3.7
+sentry-sdk==2.58.0
+semantic-version==2.10.0
+scipy==1.13.1
+rpds-py==0.27.1
+python-multipart==0.0.20
+pyparsing==3.3.2
+pynvml==13.0.1
+pydantic_core==2.46.3
+pycryptodomex==3.23.0
+pyasn1==0.6.3
+grpcio==1.80.0
+platformdirs==4.4.0
+pexpect==4.9.0
+parso==0.8.6
+orjson==3.11.5
+opencv-python==4.11.0.86
+stack-data==0.6.3
+omegaconf==2.2.1
+nvidia-nvtx-cu12==12.1.105
+nvidia-nvjitlink-cu12==12.9.86
+nvidia-nccl-cu12==2.19.3
+nvidia-curand-cu12==10.3.2.106
+nvidia-cufft-cu12==11.0.2.54
+nvidia-cuda-runtime-cu12==12.1.105
+nvidia-cuda-nvrtc-cu12==12.1.105
+nvidia-cuda-cupti-cu12==12.1.105
+nvidia-cublas-cu12==12.1.3.1
+ninja==1.13.0
+narwhals==2.20.0
+lxml==6.1.0
+kiwisolver==1.4.7
+joblib==1.5.3
+jiter==0.14.0
+imageio-ffmpeg==0.6.0
+ImageIO==2.37.2
+h5py==3.14.0
+h11==0.16.0
+fonttools==4.60.2
+ffmpy==1.0.0
+fast_kinematics==0.2.2
+executing==2.2.1
+exceptiongroup==1.3.1
+docker-pycreds==0.4.0
+distro==1.9.0
+decorator==5.2.1
+dacite==1.9.2
+cycler==0.12.1
+contourpy==1.3.0
+cloudpickle==3.1.2
+click==8.1.8
+asttokens==3.0.1
+annotated-types==0.7.0
+annotated-doc==0.0.4
+aiofiles==25.1.0
+absl-py==2.3.1
+uvicorn==0.39.0
+tiktoken==0.8.0
+scikit-learn==1.6.1
+sapien==3.0.0b1
+referencing==0.36.2
+pydantic==2.13.3
+pyasn1_modules==0.4.2
+prompt_toolkit==3.0.52
+nvidia-cusparse-cu12==12.1.0.106
+nvidia-cudnn-cu12==8.9.2.26
+nltk==3.9.2
+mujoco==2.3.7
+mdit-py-plugins==0.3.3
+matplotlib-inline==0.2.1
+matplotlib==3.7.5
+linkify-it-py==2.0.3
+jedi==0.19.2
+huggingface-hub==0.26.2
+httpcore==1.0.9
+gymnasium==0.29.1
+google-pasta==0.2.0
+gitdb==4.0.12
+cryptography==47.0.0
+blobfile==3.0.0
+anyio==4.12.1
+llmbc==0.0.0
+tyro==0.9.1
+toppra==0.6.7
+starlette==0.49.3
+seaborn==0.13.2
+rouge_score==0.1.2
+pynndescent==0.6.0
+nvidia-cusolver-cu12==11.4.5.107
+jsonschema-specifications==2025.9.1
+ipython==8.18.1
+httpx==0.28.1
+gymnasium-robotics==1.2.0
+google-auth==2.49.2
+GitPython==3.1.47
+diffusers==0.31.0
+wandb==0.18.6
+umap-learn==0.5.9.post2
+requests-oauthlib==2.0.0
+openai==2.8.1
+mplib==0.1.1
+jsonschema==4.25.1
+gradio_client==0.2.9
+google-genai==1.47.0
+fastapi==0.128.8
+torchvision==0.17.2
+torchaudio==2.2.2
+sentence-transformers==3.2.1
+pytorch-seed==0.2.0
+deepspeed==0.16.1
+bitsandbytes==0.45.0
+altair==6.0.0
+timm==1.0.22
+gradio==3.36.1
+evaluate==0.4.3
+arm_pytorch_utilities==0.5.0
+tensorflow-io-gcs-filesystem==0.37.1
+Markdown==3.9
+pybullet==3.2.6
+wrapt==1.14.2
+tensorflow-probability==0.23.0
+parse==1.19.1
+dm-tree==0.1.8
+termcolor==3.1.0
+requests==2.32.0
+cffi==2.0.0
+opt_einsum==3.4.0
+labmaze==1.0.6
+typing_extensions==4.15.0
+importlib-resources==5.13.0
+importlib-metadata==5.2.0
+FLAML==2.3.6
+dm-env==1.6
+diskcache==5.6.3
+Cython==0.29.37
+mujoco-py==2.1.2.14
+jaxlib==0.4.30
+dm-control==1.0.14
+cmudict==1.0.13
+syllables==1.0.9
+metaworld==2.0.0
+jax==0.4.30
+highway-env==1.9.1
+gym_bandits==0.0.2
+d4rl==1.1
+pyautogen==0.1.0
+protobuf==4.25.9
+tf-agents==0.19.0
+oauthlib==3.3.1
+pytorch-kinematics==0.7.5
+pygame==2.6.1
+gin-config==0.5.0
+mani_skill==3.0.0b20
+tifffile==2024.8.30
+Shapely==1.8.4
+PyWavelets==1.6.0
+Werkzeug==3.1.8
+astunparse==1.6.3
+scikit-image==0.19.3
+pymunk==6.2.1
+ml-dtypes==0.3.2
+gast==0.7.0
+stable-baselines3==2.2.1
+keras==2.15.0
+libclang==18.1.1
+flatbuffers==25.12.19
+gym==0.23.1
+mjrl==1.0.0
+llfbench==0.1.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.0.0
+inflect==7.3.1
+jaraco.collections==5.1.0
+jaraco.context==5.3.0
+jaraco.functools==4.0.1
+jaraco.text==3.12.1
+more-itertools==10.3.0
+packaging==24.2
+platformdirs==4.2.2
+tomli==2.0.1
+typeguard==4.3.0
+typing_extensions==4.12.2
+wheel==0.45.1
+zipp==3.19.2

2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083749-n45frqzz/files/wandb-metadata.json

@@ -0,0 +1,56 @@
+{
+  "os": "Linux-5.15.0-140-generic-x86_64-with-glibc2.35",
+  "python": "3.9.25",
+  "startedAt": "2026-04-29T08:37:49.387444Z",
+  "args": [
+    "--config-path",
+    "config/mix_data",
+    "--config-name",
+    "llmdp_sweep-v2_mix(csp10).yaml",
+    "training.seed=45"
+  ],
+  "program": "/workspace/LLM-BC/./train.py",
+  "codePath": "train.py",
+  "git": {
+    "remote": "https://github.com/CHYang25/LLM-BC.git",
+    "commit": "284b3dcd20dcb4b4fca100fc3721ccc07e7bb4ea"
+  },
+  "email": "chris920325@gmail.com",
+  "root": "/workspace/LLM-BC/data/outputs/2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2",
+  "host": "d8640274e028",
+  "username": "root",
+  "executable": "/venv/llm-bc/bin/python3",
+  "codePathLocal": "train.py",
+  "cpu_count": 36,
+  "cpu_count_logical": 36,
+  "gpu": "NVIDIA GeForce RTX 4090",
+  "gpu_count": 2,
+  "disk": {
+    "/": {
+      "total": "569083166720",
+      "used": "375286693888"
+    }
+  },
+  "memory": {
+    "total": "135036993536"
+  },
+  "cpu": {
+    "count": 36,
+    "countLogical": 36
+  },
+  "gpu_nvidia": [
+    {
+      "name": "NVIDIA GeForce RTX 4090",
+      "memoryTotal": "25757220864",
+      "cudaCores": 16384,
+      "architecture": "Ada"
+    },
+    {
+      "name": "NVIDIA GeForce RTX 4090",
+      "memoryTotal": "25757220864",
+      "cudaCores": 16384,
+      "architecture": "Ada"
+    }
+  ],
+  "cudaVersion": "12.4"
+}

2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083749-n45frqzz/files/wandb-summary.json

@@ -0,0 +1 @@
+{"lr":4.4e-06,"global_step":337,"_wandb":{"runtime":158},"train_loss_llm":0,"train_loss":0.7433826327323914,"train_loss_dp":0.7433826327323914,"_timestamp":1.7774519382402642e+09,"epoch":0,"_runtime":158.236138658,"_step":337}

2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/run-20260429_083749-n45frqzz/run-n45frqzz.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a9efb6a5be79c2fc5b3dfa4a3deadeab54ea687711f5c76f6a52638ce2b8bd6
+size 262144

2026.04.29/08.36.55_train_llm_diffusion_unet_lowdim_sweep-v2/wandb/wandb-resume.json

@@ -0,0 +1 @@
+{"run_id": "n45frqzz"}

2026.04.29/08.39.33_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/config.yaml

@@ -0,0 +1,193 @@
+name: train_llm_diffusion_unet_lowdim
+_target_: llmbc.workspace.train_llm_diffusion_unet_lowdim_workspace.TrainLLMDiffusionUnetLowdimWorkspace
+obs_dim: ${task.obs_dim}
+action_dim: ${task.action_dim}
+task_name: ${task.name}
+exp_name: default
+model_name: ${llm.name}
+horizon: 1
+n_obs_steps: 1
+n_action_steps: 1
+n_latency_steps: 0
+past_action_visible: false
+keypoint_visible_rate: 1.0
+obs_as_local_cond: false
+obs_as_global_cond: true
+pred_action_steps_only: false
+llm_orig_expert_feedback: true
+llm_do_sample: false
+policy:
+  _target_: llmbc.policy.llm_diffusion_unet_lowdim_policy.LLMDiffusionUnetLowdimPolicy
+  model:
+    _target_: llmbc.model.diffusion.conditional_unet1d.ConditionalUnet1D
+    input_dim: '${eval: ${task.action_dim} if ${obs_as_local_cond} or ${obs_as_global_cond}
+      else ${task.obs_dim} + ${task.action_dim}}'
+    local_cond_dim: '${eval: ${task.obs_dim} if ${obs_as_local_cond} else None}'
+    global_cond_dim: '${eval: ${task.obs_dim}*${n_obs_steps} if ${obs_as_global_cond}
+      else None}'
+    diffusion_step_embed_dim: 256
+    down_dims:
+    - 256
+    - 512
+    - 1024
+    kernel_size: 5
+    n_groups: 8
+    cond_predict_scale: true
+  noise_scheduler:
+    _target_: diffusers.schedulers.scheduling_ddpm.DDPMScheduler
+    num_train_timesteps: 100
+    beta_start: 0.0001
+    beta_end: 0.02
+    beta_schedule: squaredcos_cap_v2
+    variance_type: fixed_small
+    clip_sample: true
+    prediction_type: epsilon
+  horizon: ${horizon}
+  obs_dim: ${obs_dim}
+  action_dim: ${action_dim}
+  n_action_steps: ${eval:'${n_action_steps}+${n_latency_steps}'}
+  n_obs_steps: ${n_obs_steps}
+  num_inference_steps: 100
+  obs_as_local_cond: ${obs_as_local_cond}
+  obs_as_global_cond: ${obs_as_global_cond}
+  pred_action_steps_only: ${pred_action_steps_only}
+  oa_step_convention: true
+  llm_discriminator:
+    _target_: llmbc.discriminator.llm_ce_discriminator.LLMCEDiscriminator
+    task_id: ${task_name}
+    llm_translator:
+      _target_: llmbc.translator.llm_translator.LLMTranslator
+      cfg: ${llm}
+      obs_dim: ${obs_dim}
+      action_dim: ${action_dim}
+      horizon: ${horizon}
+      n_obs_steps: ${n_obs_steps}
+      n_action_steps: ${eval:'${n_action_steps}+${n_latency_steps}'}
+  loss_dp_weight: 1.0
+  loss_llm_weight: 0.001
+  normalize_llm_loss: true
+  reweight_llm_loss: true
+ema:
+  _target_: llmbc.model.diffusion.ema_model.EMAModel
+  update_after_step: 0
+  inv_gamma: 1.0
+  power: 0.75
+  min_value: 0.0
+  max_value: 0.9999
+dataloader:
+  batch_size: 16
+  num_workers: 0
+  shuffle: true
+  pin_memory: false
+  persistent_workers: false
+val_dataloader:
+  batch_size: 16
+  num_workers: 0
+  shuffle: false
+  pin_memory: false
+  persistent_workers: false
+optimizer:
+  _target_: torch.optim.AdamW
+  lr: 0.0001
+  betas:
+  - 0.95
+  - 0.999
+  eps: 1.0e-08
+  weight_decay: 1.0e-06
+training:
+  device: cuda:0
+  seed: 46
+  debug: false
+  resume: true
+  lr_scheduler: cosine
+  lr_warmup_steps: 500
+  num_epochs: 1001
+  gradient_accumulate_every: 16
+  use_ema: true
+  rollout_every: 5
+  checkpoint_every: 5
+  val_every: 1
+  sample_every: 5
+  max_train_steps: null
+  max_val_steps: null
+  tqdm_interval_sec: 1.0
+logging:
+  project: ${task.name}-training-mix
+  resume: true
+  mode: online
+  name: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
+  tags:
+  - ${name}
+  - ${task_name}
+  - ${exp_name}
+  id: null
+  group: null
+checkpoint:
+  topk:
+    monitor_key: test_success_rate
+    mode: max
+    k: 5
+    format_str: epoch={epoch:04d}-test_success_rate={test_success_rate:.3f}.ckpt
+  save_last_ckpt: true
+  save_last_snapshot: false
+multi_run:
+  run_dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+  wandb_name_base: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
+task:
+  name: sweep-v2
+  obs_dim: 9
+  action_dim: 4
+  env_runner:
+    _target_: llmbc.env_runner.metaworld_lowdim_runner.MetaworldLowdimRunner
+    env_name: llf-metaworld-sweep-v2
+    n_train: 10
+    n_test: 50
+    n_envs: 10
+    max_steps: 30
+    n_obs_steps: ${n_obs_steps}
+    n_action_steps: ${n_action_steps}
+    instruction_type: b
+    feedback_type:
+    - hp
+    - hn
+    - fp
+    visual: false
+    discount: 0.9
+  dataset:
+    _target_: llmbc.dataset.metaworld_lowdim_dataset.MetaworldLowdimDataset
+    data_path: datasets/sweep-v2-mix(csp10).pt
+    data_path2: datasets/sweep-v2.pt
+    horizon: ${horizon}
+    pad_before: ${eval:'${n_obs_steps}-1'}
+    pad_after: ${eval:'${n_action_steps}-1'}
+    obs_eef_target: true
+    use_manual_normalizer: false
+    val_ratio: 0.1
+    dummy_normalizer: false
+  instructor:
+    _target_: llmbc.translator.instructor.metaworld_instructor.sweep_v2_instructor.SweepV2Instructor
+llm:
+  name: HuggingFaceTB/SmolLM2-135M-Instruct
+  model_name: SmolLM2-135M-Instruct
+  config_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig
+  causal_lm_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM
+  use_quantization: false
+  use_joint_mlp_projector: true
+  llm_mode: ete-finetuned
+  finetune_mode: orig
+  checkpoint: data/outputs/2025.11.01/23.03.01_train_llm_lowdim_sweep-v2/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-sweep-v2/checkpoint-5690
+  max_length: 100
+  lora_config:
+    r: 32
+    lora_alpha: 64
+    lora_dropout: 0.05
+    bias: none
+    task_type: CAUSAL_LM
+  prompter:
+    _target_: llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter
+    use_joint_mlp_projector: true
+  hydra:
+    job:
+      override_dirname: ${model_name}
+    run:
+      dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${model_name}

2026.04.29/08.39.33_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/hydra.yaml

@@ -0,0 +1,155 @@
+hydra:
+  run:
+    dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+  sweep:
+    dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+    subdir: ${hydra.job.num}
+  launcher:
+    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
+  sweeper:
+    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
+    max_batch_size: null
+    params: null
+  help:
+    app_name: ${hydra.job.name}
+    header: '${hydra.help.app_name} is powered by Hydra.
+
+      '
+    footer: 'Powered by Hydra (https://hydra.cc)
+
+      Use --hydra-help to view Hydra specific help
+
+      '
+    template: '${hydra.help.header}
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (group=option)
+
+
+      $APP_CONFIG_GROUPS
+
+
+      == Config ==
+
+      Override anything in the config (foo.bar=value)
+
+
+      $CONFIG
+
+
+      ${hydra.help.footer}
+
+      '
+  hydra_help:
+    template: 'Hydra (${hydra.runtime.version})
+
+      See https://hydra.cc for more info.
+
+
+      == Flags ==
+
+      $FLAGS_HELP
+
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
+      to command line)
+
+
+      $HYDRA_CONFIG_GROUPS
+
+
+      Use ''--cfg hydra'' to Show the Hydra config.
+
+      '
+    hydra_help: ???
+  hydra_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][HYDRA] %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+    root:
+      level: INFO
+      handlers:
+      - console
+    loggers:
+      logging_example:
+        level: DEBUG
+    disable_existing_loggers: false
+  job_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+      file:
+        class: logging.FileHandler
+        formatter: simple
+        filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
+    root:
+      level: INFO
+      handlers:
+      - console
+      - file
+    disable_existing_loggers: false
+  env: {}
+  mode: RUN
+  searchpath: []
+  callbacks: {}
+  output_subdir: .hydra
+  overrides:
+    hydra:
+    - hydra.mode=RUN
+    task:
+    - training.seed=46
+  job:
+    name: train
+    chdir: null
+    override_dirname: training.seed=46
+    id: ???
+    num: ???
+    config_name: llmdp_sweep-v2_mix(csp10).yaml
+    env_set: {}
+    env_copy: []
+    config:
+      override_dirname:
+        kv_sep: '='
+        item_sep: ','
+        exclude_keys: []
+  runtime:
+    version: 1.2.0
+    version_base: '1.2'
+    cwd: /workspace/LLM-BC
+    config_sources:
+    - path: hydra.conf
+      schema: pkg
+      provider: hydra
+    - path: /workspace/LLM-BC/config/mix_data
+      schema: file
+      provider: main
+    - path: ''
+      schema: structured
+      provider: schema
+    output_dir: /workspace/LLM-BC/data/outputs/2026.04.29/08.39.33_train_llm_diffusion_unet_lowdim_sweep-v2
+    choices:
+      hydra/env: default
+      hydra/callbacks: null
+      hydra/job_logging: default
+      hydra/hydra_logging: default
+      hydra/hydra_help: default
+      hydra/help: default
+      hydra/sweeper: basic
+      hydra/launcher: basic
+      hydra/output: default
+  verbose: false

2026.04.29/08.39.33_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/overrides.yaml

@@ -0,0 +1 @@
+- training.seed=46

2026.04.29/08.39.33_train_llm_diffusion_unet_lowdim_sweep-v2/train.log

@@ -0,0 +1,6 @@
+[2026-04-29 08:39:35,247][numexpr.utils][INFO] - Note: NumExpr detected 36 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 16.
+[2026-04-29 08:39:35,247][numexpr.utils][INFO] - NumExpr defaulting to 16 threads.
+[2026-04-29 08:39:37,686][datasets][INFO] - PyTorch version 2.2.2 available.
+[2026-04-29 08:39:37,686][datasets][INFO] - TensorFlow version 2.15.1 available.
+[2026-04-29 08:39:37,687][datasets][INFO] - JAX version 0.4.30 available.
+[2026-04-29 08:39:39,346][llmbc.model.diffusion.conditional_unet1d][INFO] - number of parameters: 6.501479e+07

2026.04.29/08.43.27_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/config.yaml

@@ -0,0 +1,193 @@
+name: train_llm_diffusion_unet_lowdim
+_target_: llmbc.workspace.train_llm_diffusion_unet_lowdim_workspace.TrainLLMDiffusionUnetLowdimWorkspace
+obs_dim: ${task.obs_dim}
+action_dim: ${task.action_dim}
+task_name: ${task.name}
+exp_name: default
+model_name: ${llm.name}
+horizon: 1
+n_obs_steps: 1
+n_action_steps: 1
+n_latency_steps: 0
+past_action_visible: false
+keypoint_visible_rate: 1.0
+obs_as_local_cond: false
+obs_as_global_cond: true
+pred_action_steps_only: false
+llm_orig_expert_feedback: true
+llm_do_sample: false
+policy:
+  _target_: llmbc.policy.llm_diffusion_unet_lowdim_policy.LLMDiffusionUnetLowdimPolicy
+  model:
+    _target_: llmbc.model.diffusion.conditional_unet1d.ConditionalUnet1D
+    input_dim: '${eval: ${task.action_dim} if ${obs_as_local_cond} or ${obs_as_global_cond}
+      else ${task.obs_dim} + ${task.action_dim}}'
+    local_cond_dim: '${eval: ${task.obs_dim} if ${obs_as_local_cond} else None}'
+    global_cond_dim: '${eval: ${task.obs_dim}*${n_obs_steps} if ${obs_as_global_cond}
+      else None}'
+    diffusion_step_embed_dim: 256
+    down_dims:
+    - 256
+    - 512
+    - 1024
+    kernel_size: 5
+    n_groups: 8
+    cond_predict_scale: true
+  noise_scheduler:
+    _target_: diffusers.schedulers.scheduling_ddpm.DDPMScheduler
+    num_train_timesteps: 100
+    beta_start: 0.0001
+    beta_end: 0.02
+    beta_schedule: squaredcos_cap_v2
+    variance_type: fixed_small
+    clip_sample: true
+    prediction_type: epsilon
+  horizon: ${horizon}
+  obs_dim: ${obs_dim}
+  action_dim: ${action_dim}
+  n_action_steps: ${eval:'${n_action_steps}+${n_latency_steps}'}
+  n_obs_steps: ${n_obs_steps}
+  num_inference_steps: 100
+  obs_as_local_cond: ${obs_as_local_cond}
+  obs_as_global_cond: ${obs_as_global_cond}
+  pred_action_steps_only: ${pred_action_steps_only}
+  oa_step_convention: true
+  llm_discriminator:
+    _target_: llmbc.discriminator.llm_ce_discriminator.LLMCEDiscriminator
+    task_id: ${task_name}
+    llm_translator:
+      _target_: llmbc.translator.llm_translator.LLMTranslator
+      cfg: ${llm}
+      obs_dim: ${obs_dim}
+      action_dim: ${action_dim}
+      horizon: ${horizon}
+      n_obs_steps: ${n_obs_steps}
+      n_action_steps: ${eval:'${n_action_steps}+${n_latency_steps}'}
+  loss_dp_weight: 1.0
+  loss_llm_weight: 0.001
+  normalize_llm_loss: true
+  reweight_llm_loss: true
+ema:
+  _target_: llmbc.model.diffusion.ema_model.EMAModel
+  update_after_step: 0
+  inv_gamma: 1.0
+  power: 0.75
+  min_value: 0.0
+  max_value: 0.9999
+dataloader:
+  batch_size: 16
+  num_workers: 0
+  shuffle: true
+  pin_memory: false
+  persistent_workers: false
+val_dataloader:
+  batch_size: 16
+  num_workers: 0
+  shuffle: false
+  pin_memory: false
+  persistent_workers: false
+optimizer:
+  _target_: torch.optim.AdamW
+  lr: 0.0001
+  betas:
+  - 0.95
+  - 0.999
+  eps: 1.0e-08
+  weight_decay: 1.0e-06
+training:
+  device: cuda:0
+  seed: 44
+  debug: false
+  resume: true
+  lr_scheduler: cosine
+  lr_warmup_steps: 500
+  num_epochs: 1001
+  gradient_accumulate_every: 16
+  use_ema: true
+  rollout_every: 5
+  checkpoint_every: 5
+  val_every: 1
+  sample_every: 5
+  max_train_steps: null
+  max_val_steps: null
+  tqdm_interval_sec: 1.0
+logging:
+  project: ${task.name}-training-mix
+  resume: true
+  mode: online
+  name: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
+  tags:
+  - ${name}
+  - ${task_name}
+  - ${exp_name}
+  id: null
+  group: null
+checkpoint:
+  topk:
+    monitor_key: test_success_rate
+    mode: max
+    k: 5
+    format_str: epoch={epoch:04d}-test_success_rate={test_success_rate:.3f}.ckpt
+  save_last_ckpt: true
+  save_last_snapshot: false
+multi_run:
+  run_dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+  wandb_name_base: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
+task:
+  name: sweep-v2
+  obs_dim: 9
+  action_dim: 4
+  env_runner:
+    _target_: llmbc.env_runner.metaworld_lowdim_runner.MetaworldLowdimRunner
+    env_name: llf-metaworld-sweep-v2
+    n_train: 10
+    n_test: 50
+    n_envs: 10
+    max_steps: 30
+    n_obs_steps: ${n_obs_steps}
+    n_action_steps: ${n_action_steps}
+    instruction_type: b
+    feedback_type:
+    - hp
+    - hn
+    - fp
+    visual: false
+    discount: 0.9
+  dataset:
+    _target_: llmbc.dataset.metaworld_lowdim_dataset.MetaworldLowdimDataset
+    data_path: datasets/sweep-v2-mix(csp10).pt
+    data_path2: datasets/sweep-v2.pt
+    horizon: ${horizon}
+    pad_before: ${eval:'${n_obs_steps}-1'}
+    pad_after: ${eval:'${n_action_steps}-1'}
+    obs_eef_target: true
+    use_manual_normalizer: false
+    val_ratio: 0.1
+    dummy_normalizer: false
+  instructor:
+    _target_: llmbc.translator.instructor.metaworld_instructor.sweep_v2_instructor.SweepV2Instructor
+llm:
+  name: HuggingFaceTB/SmolLM2-135M-Instruct
+  model_name: SmolLM2-135M-Instruct
+  config_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig
+  causal_lm_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM
+  use_quantization: false
+  use_joint_mlp_projector: true
+  llm_mode: ete-finetuned
+  finetune_mode: orig
+  checkpoint: data/outputs/2025.11.01/23.03.01_train_llm_lowdim_sweep-v2/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-sweep-v2/checkpoint-5690
+  max_length: 100
+  lora_config:
+    r: 32
+    lora_alpha: 64
+    lora_dropout: 0.05
+    bias: none
+    task_type: CAUSAL_LM
+  prompter:
+    _target_: llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter
+    use_joint_mlp_projector: true
+  hydra:
+    job:
+      override_dirname: ${model_name}
+    run:
+      dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${model_name}

2026.04.29/08.43.27_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/hydra.yaml

@@ -0,0 +1,155 @@
+hydra:
+  run:
+    dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+  sweep:
+    dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
+    subdir: ${hydra.job.num}
+  launcher:
+    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
+  sweeper:
+    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
+    max_batch_size: null
+    params: null
+  help:
+    app_name: ${hydra.job.name}
+    header: '${hydra.help.app_name} is powered by Hydra.
+
+      '
+    footer: 'Powered by Hydra (https://hydra.cc)
+
+      Use --hydra-help to view Hydra specific help
+
+      '
+    template: '${hydra.help.header}
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (group=option)
+
+
+      $APP_CONFIG_GROUPS
+
+
+      == Config ==
+
+      Override anything in the config (foo.bar=value)
+
+
+      $CONFIG
+
+
+      ${hydra.help.footer}
+
+      '
+  hydra_help:
+    template: 'Hydra (${hydra.runtime.version})
+
+      See https://hydra.cc for more info.
+
+
+      == Flags ==
+
+      $FLAGS_HELP
+
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
+      to command line)
+
+
+      $HYDRA_CONFIG_GROUPS
+
+
+      Use ''--cfg hydra'' to Show the Hydra config.
+
+      '
+    hydra_help: ???
+  hydra_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][HYDRA] %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+    root:
+      level: INFO
+      handlers:
+      - console
+    loggers:
+      logging_example:
+        level: DEBUG
+    disable_existing_loggers: false
+  job_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+      file:
+        class: logging.FileHandler
+        formatter: simple
+        filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
+    root:
+      level: INFO
+      handlers:
+      - console
+      - file
+    disable_existing_loggers: false
+  env: {}
+  mode: RUN
+  searchpath: []
+  callbacks: {}
+  output_subdir: .hydra
+  overrides:
+    hydra:
+    - hydra.mode=RUN
+    task:
+    - training.seed=44
+  job:
+    name: train
+    chdir: null
+    override_dirname: training.seed=44
+    id: ???
+    num: ???
+    config_name: llmdp_sweep-v2_mix(csp10).yaml
+    env_set: {}
+    env_copy: []
+    config:
+      override_dirname:
+        kv_sep: '='
+        item_sep: ','
+        exclude_keys: []
+  runtime:
+    version: 1.2.0
+    version_base: '1.2'
+    cwd: /workspace/LLM-BC
+    config_sources:
+    - path: hydra.conf
+      schema: pkg
+      provider: hydra
+    - path: /workspace/LLM-BC/config/mix_data
+      schema: file
+      provider: main
+    - path: ''
+      schema: structured
+      provider: schema
+    output_dir: /workspace/LLM-BC/data/outputs/2026.04.29/08.43.27_train_llm_diffusion_unet_lowdim_sweep-v2
+    choices:
+      hydra/env: default
+      hydra/callbacks: null
+      hydra/job_logging: default
+      hydra/hydra_logging: default
+      hydra/hydra_help: default
+      hydra/help: default
+      hydra/sweeper: basic
+      hydra/launcher: basic
+      hydra/output: default
+  verbose: false

2026.04.29/08.43.27_train_llm_diffusion_unet_lowdim_sweep-v2/.hydra/overrides.yaml

@@ -0,0 +1 @@
+- training.seed=44

2026.04.29/08.43.27_train_llm_diffusion_unet_lowdim_sweep-v2/checkpoints/epoch=0435-test_success_rate=0.980.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb7bf1184a4c1494b6c24cff86bbdb43e01c0d9f8c9ada9df89db44bb81b7f00
+size 1040498994

2026.04.29/08.43.27_train_llm_diffusion_unet_lowdim_sweep-v2/checkpoints/epoch=0520-test_success_rate=0.960.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba0e3688d10151615be6582c01b69d2dcaeba9411d257eec6fb50ebed6b3c770
+size 1040498994