Upload folder using huggingface_hub
Browse files- checkpoint-200/embodiment_id.json +2 -1
- checkpoint-200/experiment_cfg/conf.yaml +1 -1
- checkpoint-200/experiment_cfg/config.yaml +1 -1
- checkpoint-200/experiment_cfg/final_processor_config.json +1 -1
- checkpoint-200/global_step200/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +2 -2
- checkpoint-200/global_step200/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +2 -2
- checkpoint-200/global_step200/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt +2 -2
- checkpoint-200/global_step200/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt +2 -2
- checkpoint-200/global_step200/mp_rank_00_model_states.pt +1 -1
- checkpoint-200/model-00001-of-00002.safetensors +1 -1
- checkpoint-200/model-00002-of-00002.safetensors +1 -1
- checkpoint-200/rng_state_0.pth +2 -2
- checkpoint-200/rng_state_1.pth +2 -2
- checkpoint-200/rng_state_2.pth +2 -2
- checkpoint-200/rng_state_3.pth +2 -2
- checkpoint-200/trainer_state.json +41 -41
- checkpoint-200/training_args.bin +1 -1
checkpoint-200/embodiment_id.json
CHANGED
|
@@ -52,5 +52,6 @@
|
|
| 52 |
"libero_sim": 2,
|
| 53 |
"droid_sim": 3,
|
| 54 |
"unitree_g1_sonic": 11,
|
| 55 |
-
"new_embodiment": 10
|
|
|
|
| 56 |
}
|
|
|
|
| 52 |
"libero_sim": 2,
|
| 53 |
"droid_sim": 3,
|
| 54 |
"unitree_g1_sonic": 11,
|
| 55 |
+
"new_embodiment": 10,
|
| 56 |
+
"robocasa_gr1_tabletop": 10
|
| 57 |
}
|
checkpoint-200/experiment_cfg/conf.yaml
CHANGED
|
@@ -233,7 +233,7 @@ training:
|
|
| 233 |
transformers_access_token: null
|
| 234 |
use_ddp: false
|
| 235 |
ddp_bucket_cap_mb: 100
|
| 236 |
-
num_gpus:
|
| 237 |
dataloader_num_workers: 6
|
| 238 |
remove_unused_columns: false
|
| 239 |
use_wandb: true
|
|
|
|
| 233 |
transformers_access_token: null
|
| 234 |
use_ddp: false
|
| 235 |
ddp_bucket_cap_mb: 100
|
| 236 |
+
num_gpus: 4
|
| 237 |
dataloader_num_workers: 6
|
| 238 |
remove_unused_columns: false
|
| 239 |
use_wandb: true
|
checkpoint-200/experiment_cfg/config.yaml
CHANGED
|
@@ -245,7 +245,7 @@ training: !!python/object:gr00t.configs.training.training_config.TrainingConfig
|
|
| 245 |
max_grad_norm: 1.0
|
| 246 |
max_retries: 3
|
| 247 |
max_steps: 200
|
| 248 |
-
num_gpus:
|
| 249 |
open_loop_eval_plot_indices: null
|
| 250 |
open_loop_eval_steps_per_traj: 100
|
| 251 |
open_loop_eval_traj_ids:
|
|
|
|
| 245 |
max_grad_norm: 1.0
|
| 246 |
max_retries: 3
|
| 247 |
max_steps: 200
|
| 248 |
+
num_gpus: 4
|
| 249 |
open_loop_eval_plot_indices: null
|
| 250 |
open_loop_eval_steps_per_traj: 100
|
| 251 |
open_loop_eval_traj_ids:
|
checkpoint-200/experiment_cfg/final_processor_config.json
CHANGED
|
@@ -21,7 +21,7 @@
|
|
| 21 |
"random_rotation_angle": "0",
|
| 22 |
"color_jitter_params": "{'brightness': 0.3, 'contrast': 0.4, 'saturation': 0.5, 'hue': 0.08}",
|
| 23 |
"processor": "Qwen3VLProcessor:\n- image_processor: Qwen2VLImageProcessorFast {\n \"crop_size\": null,\n \"data_format\": \"channels_first\",\n \"default_to_square\": true,\n \"device\": null,\n \"disable_grouping\": null,\n \"do_center_crop\": null,\n \"do_convert_rgb\": true,\n \"do_normalize\": true,\n \"do_pad\": null,\n \"do_rescale\": true,\n \"do_resize\": true,\n \"image_mean\": [\n 0.5,\n 0.5,\n 0.5\n ],\n \"image_processor_type\": \"Qwen2VLImageProcessorFast\",\n \"image_std\": [\n 0.5,\n 0.5,\n 0.5\n ],\n \"input_data_format\": null,\n \"max_pixels\": null,\n \"merge_size\": 2,\n \"min_pixels\": null,\n \"pad_size\": null,\n \"patch_size\": 16,\n \"processor_class\": \"Qwen3VLProcessor\",\n \"resample\": 3,\n \"rescale_factor\": 0.00392156862745098,\n \"return_tensors\": null,\n \"size\": {\n \"longest_edge\": 16777216,\n \"shortest_edge\": 65536\n },\n \"temporal_patch_size\": 2\n}\n\n- tokenizer: Qwen2TokenizerFast(name_or_path='nvidia/Cosmos-Reason2-2B', vocab_size=151643, model_max_length=262144, is_fast=True, padding_side='left', truncation_side='right', special_tokens={'eos_token': '<|im_end|>', 'pad_token': '<|endoftext|>', 'additional_special_tokens': ['<|im_start|>', '<|im_end|>', '<|object_ref_start|>', '<|object_ref_end|>', '<|box_start|>', '<|box_end|>', '<|quad_start|>', '<|quad_end|>', '<|vision_start|>', '<|vision_end|>', '<|vision_pad|>', '<|image_pad|>', '<|video_pad|>']}, clean_up_tokenization_spaces=False, added_tokens_decoder={\n\t151643: AddedToken(\"<|endoftext|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151644: AddedToken(\"<|im_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151645: AddedToken(\"<|im_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151646: AddedToken(\"<|object_ref_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151647: AddedToken(\"<|object_ref_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151648: AddedToken(\"<|box_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151649: AddedToken(\"<|box_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151650: AddedToken(\"<|quad_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151651: AddedToken(\"<|quad_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151652: AddedToken(\"<|vision_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151653: AddedToken(\"<|vision_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151654: AddedToken(\"<|vision_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151655: AddedToken(\"<|image_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151656: AddedToken(\"<|video_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151657: AddedToken(\"<tool_call>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151658: AddedToken(\"</tool_call>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151659: AddedToken(\"<|fim_prefix|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151660: AddedToken(\"<|fim_middle|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151661: AddedToken(\"<|fim_suffix|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151662: AddedToken(\"<|fim_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151663: AddedToken(\"<|repo_name|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151664: AddedToken(\"<|file_sep|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151665: AddedToken(\"<tool_response>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151666: AddedToken(\"</tool_response>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151667: AddedToken(\"<think>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151668: AddedToken(\"</think>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n}\n)\n- video_processor: Qwen3VLVideoProcessor {\n \"crop_size\": null,\n \"data_format\": \"channels_first\",\n \"default_to_square\": true,\n \"device\": null,\n \"do_center_crop\": null,\n \"do_convert_rgb\": true,\n \"do_normalize\": true,\n \"do_rescale\": true,\n \"do_resize\": true,\n \"do_sample_frames\": true,\n \"fps\": 2,\n \"image_mean\": [\n 0.5,\n 0.5,\n 0.5\n ],\n \"image_std\": [\n 0.5,\n 0.5,\n 0.5\n ],\n \"input_data_format\": null,\n \"max_frames\": 768,\n \"merge_size\": 2,\n \"min_frames\": 4,\n \"num_frames\": null,\n \"pad_size\": null,\n \"patch_size\": 16,\n \"processor_class\": \"Qwen3VLProcessor\",\n \"resample\": 3,\n \"rescale_factor\": 0.00392156862745098,\n \"return_metadata\": false,\n \"size\": {\n \"longest_edge\": 25165824,\n \"shortest_edge\": 4096\n },\n \"temporal_patch_size\": 2,\n \"video_metadata\": null,\n \"video_processor_type\": \"Qwen3VLVideoProcessor\"\n}\n\n\n{\n \"processor_class\": \"Qwen3VLProcessor\"\n}\n",
|
| 24 |
-
"embodiment_id_mapping": "{'robocasa_panda_omron': 13, 'oxe_droid': 17, 'oxe_fractal': 18, 'oxe_language_table': 19, 'oxe_bridge': 20, 'unknown': 22, 'gr1_unified': 20, 'agibot': 26, 'sim_behavior_r1_pro': 23, 'xdof': 24, 'xdof_oss_data': 25, 'unitree_g1_full_body_with_waist_height_nav_cmd': 25, 'real_r1_pro_sharpa': 27, 'real_r1_pro_sharpa_add_view': 27, 'real_r1_pro_sharpa_relative_arm_joint': 26, 'real_r1_pro_sharpa_delta_eef': 26, 'real_r1_pro_sharpa_absolute_eef': 26, 'real_r1_pro_sharpa_meanstd': 26, 'real_r1_pro_sharpa_relative_eef': 26, 'real_r1_pro_sharpa_relative_eef_add_view': 26, 'real_r1_pro_sharpa_relative_eef_relative_hand': 26, 'real_r1_pro_sharpa_relative_eef_human': 26, 'real_r1_pro_sharpa_relative_eef_human_add_view': 26, 'real_r1_pro_sharpa_relative_eef_human_relative_hand': 26, 'real_r1_pro_sharpa_relative_eef_egodex': 26, 'real_r1_pro_sharpa_relative_eef_egodex_relative_hand': 26, 'real_r1_pro_sharpa_relative_eef_egodex_wrist_only': 26, 'real_r1_pro_sharpa_relative_eef_maxinsights': 26, 'real_r1_pro_sharpa_relative_eef_maxinsights_relative_hand': 26, 'real_r1_pro_sharpa_relative_eef_mecka': 26, 'real_r1_pro_sharpa_relative_eef_mecka_relative_hand': 26, 'real_g1_relative_eef_absolute_joints': 25, 'real_g1_relative_eef_absolute_joints_wrist_cam': 25, 'real_g1_relative_eef_relative_joints': 25, 'real_r1_pro_sharpa_relative_eef_relative_hand_relative_joint': 26, 'real_r1_pro_sharpa_relative_joint': 29, 'oxe_droid_relative_eef_relative_joint': 24, 'oxe_droid_relative_eef_relative_joint_swapped': 24, 'oxe_droid_relative_eef_relative_joint_upweight_z': 24, 'oxe_droid_relative_eef_relative_joint_upweight_z_swapped': 24, 'oxe_droid_relative_eef_relative_joint_3view': 24, 'oxe_droid_relative_eef_relative_joint_3view_swapped': 24, 'oxe_droid_relative_eef': 24, 'oxe_droid_joint_position_relative': 24, 'xdof_relative_eef_relative_joint': 27, 'xdof_relative_eef_relative_joint_subtask': 27, 'xdof_relative_eef': 27, 'xdof_relative_joint': 28, 'simpler_env_google': 0, 'simpler_env_widowx': 1, 'libero_sim': 2, 'droid_sim': 3, 'unitree_g1_sonic': 11, 'new_embodiment': 10}",
|
| 25 |
"shortest_image_edge": "256",
|
| 26 |
"crop_fraction": "0.95",
|
| 27 |
"statistics": "{}",
|
|
|
|
| 21 |
"random_rotation_angle": "0",
|
| 22 |
"color_jitter_params": "{'brightness': 0.3, 'contrast': 0.4, 'saturation': 0.5, 'hue': 0.08}",
|
| 23 |
"processor": "Qwen3VLProcessor:\n- image_processor: Qwen2VLImageProcessorFast {\n \"crop_size\": null,\n \"data_format\": \"channels_first\",\n \"default_to_square\": true,\n \"device\": null,\n \"disable_grouping\": null,\n \"do_center_crop\": null,\n \"do_convert_rgb\": true,\n \"do_normalize\": true,\n \"do_pad\": null,\n \"do_rescale\": true,\n \"do_resize\": true,\n \"image_mean\": [\n 0.5,\n 0.5,\n 0.5\n ],\n \"image_processor_type\": \"Qwen2VLImageProcessorFast\",\n \"image_std\": [\n 0.5,\n 0.5,\n 0.5\n ],\n \"input_data_format\": null,\n \"max_pixels\": null,\n \"merge_size\": 2,\n \"min_pixels\": null,\n \"pad_size\": null,\n \"patch_size\": 16,\n \"processor_class\": \"Qwen3VLProcessor\",\n \"resample\": 3,\n \"rescale_factor\": 0.00392156862745098,\n \"return_tensors\": null,\n \"size\": {\n \"longest_edge\": 16777216,\n \"shortest_edge\": 65536\n },\n \"temporal_patch_size\": 2\n}\n\n- tokenizer: Qwen2TokenizerFast(name_or_path='nvidia/Cosmos-Reason2-2B', vocab_size=151643, model_max_length=262144, is_fast=True, padding_side='left', truncation_side='right', special_tokens={'eos_token': '<|im_end|>', 'pad_token': '<|endoftext|>', 'additional_special_tokens': ['<|im_start|>', '<|im_end|>', '<|object_ref_start|>', '<|object_ref_end|>', '<|box_start|>', '<|box_end|>', '<|quad_start|>', '<|quad_end|>', '<|vision_start|>', '<|vision_end|>', '<|vision_pad|>', '<|image_pad|>', '<|video_pad|>']}, clean_up_tokenization_spaces=False, added_tokens_decoder={\n\t151643: AddedToken(\"<|endoftext|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151644: AddedToken(\"<|im_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151645: AddedToken(\"<|im_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151646: AddedToken(\"<|object_ref_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151647: AddedToken(\"<|object_ref_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151648: AddedToken(\"<|box_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151649: AddedToken(\"<|box_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151650: AddedToken(\"<|quad_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151651: AddedToken(\"<|quad_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151652: AddedToken(\"<|vision_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151653: AddedToken(\"<|vision_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151654: AddedToken(\"<|vision_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151655: AddedToken(\"<|image_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151656: AddedToken(\"<|video_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151657: AddedToken(\"<tool_call>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151658: AddedToken(\"</tool_call>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151659: AddedToken(\"<|fim_prefix|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151660: AddedToken(\"<|fim_middle|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151661: AddedToken(\"<|fim_suffix|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151662: AddedToken(\"<|fim_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151663: AddedToken(\"<|repo_name|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151664: AddedToken(\"<|file_sep|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151665: AddedToken(\"<tool_response>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151666: AddedToken(\"</tool_response>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151667: AddedToken(\"<think>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151668: AddedToken(\"</think>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n}\n)\n- video_processor: Qwen3VLVideoProcessor {\n \"crop_size\": null,\n \"data_format\": \"channels_first\",\n \"default_to_square\": true,\n \"device\": null,\n \"do_center_crop\": null,\n \"do_convert_rgb\": true,\n \"do_normalize\": true,\n \"do_rescale\": true,\n \"do_resize\": true,\n \"do_sample_frames\": true,\n \"fps\": 2,\n \"image_mean\": [\n 0.5,\n 0.5,\n 0.5\n ],\n \"image_std\": [\n 0.5,\n 0.5,\n 0.5\n ],\n \"input_data_format\": null,\n \"max_frames\": 768,\n \"merge_size\": 2,\n \"min_frames\": 4,\n \"num_frames\": null,\n \"pad_size\": null,\n \"patch_size\": 16,\n \"processor_class\": \"Qwen3VLProcessor\",\n \"resample\": 3,\n \"rescale_factor\": 0.00392156862745098,\n \"return_metadata\": false,\n \"size\": {\n \"longest_edge\": 25165824,\n \"shortest_edge\": 4096\n },\n \"temporal_patch_size\": 2,\n \"video_metadata\": null,\n \"video_processor_type\": \"Qwen3VLVideoProcessor\"\n}\n\n\n{\n \"processor_class\": \"Qwen3VLProcessor\"\n}\n",
|
| 24 |
+
"embodiment_id_mapping": "{'robocasa_panda_omron': 13, 'oxe_droid': 17, 'oxe_fractal': 18, 'oxe_language_table': 19, 'oxe_bridge': 20, 'unknown': 22, 'gr1_unified': 20, 'agibot': 26, 'sim_behavior_r1_pro': 23, 'xdof': 24, 'xdof_oss_data': 25, 'unitree_g1_full_body_with_waist_height_nav_cmd': 25, 'real_r1_pro_sharpa': 27, 'real_r1_pro_sharpa_add_view': 27, 'real_r1_pro_sharpa_relative_arm_joint': 26, 'real_r1_pro_sharpa_delta_eef': 26, 'real_r1_pro_sharpa_absolute_eef': 26, 'real_r1_pro_sharpa_meanstd': 26, 'real_r1_pro_sharpa_relative_eef': 26, 'real_r1_pro_sharpa_relative_eef_add_view': 26, 'real_r1_pro_sharpa_relative_eef_relative_hand': 26, 'real_r1_pro_sharpa_relative_eef_human': 26, 'real_r1_pro_sharpa_relative_eef_human_add_view': 26, 'real_r1_pro_sharpa_relative_eef_human_relative_hand': 26, 'real_r1_pro_sharpa_relative_eef_egodex': 26, 'real_r1_pro_sharpa_relative_eef_egodex_relative_hand': 26, 'real_r1_pro_sharpa_relative_eef_egodex_wrist_only': 26, 'real_r1_pro_sharpa_relative_eef_maxinsights': 26, 'real_r1_pro_sharpa_relative_eef_maxinsights_relative_hand': 26, 'real_r1_pro_sharpa_relative_eef_mecka': 26, 'real_r1_pro_sharpa_relative_eef_mecka_relative_hand': 26, 'real_g1_relative_eef_absolute_joints': 25, 'real_g1_relative_eef_absolute_joints_wrist_cam': 25, 'real_g1_relative_eef_relative_joints': 25, 'real_r1_pro_sharpa_relative_eef_relative_hand_relative_joint': 26, 'real_r1_pro_sharpa_relative_joint': 29, 'oxe_droid_relative_eef_relative_joint': 24, 'oxe_droid_relative_eef_relative_joint_swapped': 24, 'oxe_droid_relative_eef_relative_joint_upweight_z': 24, 'oxe_droid_relative_eef_relative_joint_upweight_z_swapped': 24, 'oxe_droid_relative_eef_relative_joint_3view': 24, 'oxe_droid_relative_eef_relative_joint_3view_swapped': 24, 'oxe_droid_relative_eef': 24, 'oxe_droid_joint_position_relative': 24, 'xdof_relative_eef_relative_joint': 27, 'xdof_relative_eef_relative_joint_subtask': 27, 'xdof_relative_eef': 27, 'xdof_relative_joint': 28, 'simpler_env_google': 0, 'simpler_env_widowx': 1, 'libero_sim': 2, 'droid_sim': 3, 'unitree_g1_sonic': 11, 'new_embodiment': 10, 'robocasa_gr1_tabletop': 10}",
|
| 25 |
"shortest_image_edge": "256",
|
| 26 |
"crop_fraction": "0.95",
|
| 27 |
"statistics": "{}",
|
checkpoint-200/global_step200/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c7a2988b9429d26f834daf54126c2aec1d74875fffa8ec7018acacd1062531a8
|
| 3 |
+
size 4861568625
|
checkpoint-200/global_step200/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:33bea903a22668985d9940f3352ac71c8ed2b93dc18fdc4c89069ae80a0f490d
|
| 3 |
+
size 4861568369
|
checkpoint-200/global_step200/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:883614873568ec558410c180e6686fcced8be4b959ba000192a2932896f9705a
|
| 3 |
+
size 4861566321
|
checkpoint-200/global_step200/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67324024e5bd44a4f24c5c92a32660df09152988c95688df5cbeb6a0f4396db9
|
| 3 |
+
size 4861563121
|
checkpoint-200/global_step200/mp_rank_00_model_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 9335640879
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32f823a47fe5ee3ea4a78f50f44fb842b14d50c1413fa70e3e9874af118d9666
|
| 3 |
size 9335640879
|
checkpoint-200/model-00001-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4990519232
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:780ab19548b8f3ec6892023bfb21264b0babe9df35e4fe495dc281c81f8d87c7
|
| 3 |
size 4990519232
|
checkpoint-200/model-00002-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1919980184
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a61270b555246d86a0b0483cb54f31265ae1d4f4c57acd64817cd244bfc15a6
|
| 3 |
size 1919980184
|
checkpoint-200/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:585f5e83ab2e47062abc72edf125d0506c7fd66dab9442016b9eb5677cfa6221
|
| 3 |
+
size 15429
|
checkpoint-200/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5854ff429ff758a2a53c029ba5bdd58cf6e6b042d55f9732c23e9d0270fff740
|
| 3 |
+
size 15429
|
checkpoint-200/rng_state_2.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8e18c420a8712f7aa4cbc826e57af75af90687a03861f07cca8f7b457cc73557
|
| 3 |
+
size 15429
|
checkpoint-200/rng_state_3.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe3f74e890f0cc772511802269c2817b638ba2496bab58391696237ec9e78198
|
| 3 |
+
size 15429
|
checkpoint-200/trainer_state.json
CHANGED
|
@@ -10,123 +10,123 @@
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
-
"grad_norm": 0.
|
| 14 |
"learning_rate": 9e-05,
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"grad_norm": 0.
|
| 20 |
"learning_rate": 9.944739353007344e-05,
|
| 21 |
-
"loss": 1.
|
| 22 |
"step": 20
|
| 23 |
},
|
| 24 |
{
|
| 25 |
-
"grad_norm": 0.
|
| 26 |
"learning_rate": 9.755282581475769e-05,
|
| 27 |
-
"loss": 1.
|
| 28 |
"step": 30
|
| 29 |
},
|
| 30 |
{
|
| 31 |
-
"grad_norm": 0.
|
| 32 |
"learning_rate": 9.43611409721806e-05,
|
| 33 |
-
"loss": 1.
|
| 34 |
"step": 40
|
| 35 |
},
|
| 36 |
{
|
| 37 |
-
"grad_norm": 0.
|
| 38 |
"learning_rate": 8.995939984474624e-05,
|
| 39 |
-
"loss": 1.
|
| 40 |
"step": 50
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"grad_norm": 0.
|
| 44 |
"learning_rate": 8.44676704559283e-05,
|
| 45 |
-
"loss": 1.
|
| 46 |
"step": 60
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"grad_norm": 0.
|
| 50 |
"learning_rate": 7.803575286758364e-05,
|
| 51 |
-
"loss": 1.
|
| 52 |
"step": 70
|
| 53 |
},
|
| 54 |
{
|
| 55 |
-
"grad_norm": 0.
|
| 56 |
"learning_rate": 7.083909302476453e-05,
|
| 57 |
-
"loss": 1.
|
| 58 |
"step": 80
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"grad_norm": 0.
|
| 62 |
"learning_rate": 6.307399704769099e-05,
|
| 63 |
-
"loss": 1.
|
| 64 |
"step": 90
|
| 65 |
},
|
| 66 |
{
|
| 67 |
-
"grad_norm": 0.
|
| 68 |
"learning_rate": 5.495227651252315e-05,
|
| 69 |
-
"loss": 1.
|
| 70 |
"step": 100
|
| 71 |
},
|
| 72 |
{
|
| 73 |
-
"grad_norm": 0.
|
| 74 |
"learning_rate": 4.669547078371504e-05,
|
| 75 |
-
"loss": 1.
|
| 76 |
"step": 110
|
| 77 |
},
|
| 78 |
{
|
| 79 |
-
"grad_norm": 0.
|
| 80 |
"learning_rate": 3.852880399766243e-05,
|
| 81 |
-
"loss": 1.
|
| 82 |
"step": 120
|
| 83 |
},
|
| 84 |
{
|
| 85 |
-
"grad_norm": 0.
|
| 86 |
"learning_rate": 3.0675041535377405e-05,
|
| 87 |
-
"loss": 1.
|
| 88 |
"step": 130
|
| 89 |
},
|
| 90 |
{
|
| 91 |
-
"grad_norm": 0.
|
| 92 |
"learning_rate": 2.3348413563600325e-05,
|
| 93 |
-
"loss": 1.
|
| 94 |
"step": 140
|
| 95 |
},
|
| 96 |
{
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
"learning_rate": 1.6748771394307585e-05,
|
| 99 |
-
"loss": 1.
|
| 100 |
"step": 150
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"grad_norm": 0.
|
| 104 |
"learning_rate": 1.1056136061894384e-05,
|
| 105 |
-
"loss": 1.
|
| 106 |
"step": 160
|
| 107 |
},
|
| 108 |
{
|
| 109 |
-
"grad_norm": 0.
|
| 110 |
"learning_rate": 6.425787818636131e-06,
|
| 111 |
-
"loss": 1.
|
| 112 |
"step": 170
|
| 113 |
},
|
| 114 |
{
|
| 115 |
-
"grad_norm": 0.
|
| 116 |
"learning_rate": 2.9840304941919415e-06,
|
| 117 |
-
"loss": 1.
|
| 118 |
"step": 180
|
| 119 |
},
|
| 120 |
{
|
| 121 |
-
"grad_norm": 0.
|
| 122 |
"learning_rate": 8.247462563808817e-07,
|
| 123 |
-
"loss": 1.
|
| 124 |
"step": 190
|
| 125 |
},
|
| 126 |
{
|
| 127 |
-
"grad_norm": 0.
|
| 128 |
"learning_rate": 6.834750376549792e-09,
|
| 129 |
-
"loss": 1.
|
| 130 |
"step": 200
|
| 131 |
}
|
| 132 |
],
|
|
@@ -148,7 +148,7 @@
|
|
| 148 |
}
|
| 149 |
},
|
| 150 |
"total_flos": 0.0,
|
| 151 |
-
"train_batch_size":
|
| 152 |
"trial_name": null,
|
| 153 |
"trial_params": null
|
| 154 |
}
|
|
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
+
"grad_norm": 0.23940400779247284,
|
| 14 |
"learning_rate": 9e-05,
|
| 15 |
+
"loss": 1.1936,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"grad_norm": 0.6992996335029602,
|
| 20 |
"learning_rate": 9.944739353007344e-05,
|
| 21 |
+
"loss": 1.1512,
|
| 22 |
"step": 20
|
| 23 |
},
|
| 24 |
{
|
| 25 |
+
"grad_norm": 0.2213364690542221,
|
| 26 |
"learning_rate": 9.755282581475769e-05,
|
| 27 |
+
"loss": 1.1256,
|
| 28 |
"step": 30
|
| 29 |
},
|
| 30 |
{
|
| 31 |
+
"grad_norm": 0.2311399132013321,
|
| 32 |
"learning_rate": 9.43611409721806e-05,
|
| 33 |
+
"loss": 1.0988,
|
| 34 |
"step": 40
|
| 35 |
},
|
| 36 |
{
|
| 37 |
+
"grad_norm": 0.23056164383888245,
|
| 38 |
"learning_rate": 8.995939984474624e-05,
|
| 39 |
+
"loss": 1.1029,
|
| 40 |
"step": 50
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"grad_norm": 0.28725796937942505,
|
| 44 |
"learning_rate": 8.44676704559283e-05,
|
| 45 |
+
"loss": 1.102,
|
| 46 |
"step": 60
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"grad_norm": 0.33148351311683655,
|
| 50 |
"learning_rate": 7.803575286758364e-05,
|
| 51 |
+
"loss": 1.0932,
|
| 52 |
"step": 70
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"grad_norm": 0.1882592737674713,
|
| 56 |
"learning_rate": 7.083909302476453e-05,
|
| 57 |
+
"loss": 1.0951,
|
| 58 |
"step": 80
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"grad_norm": 0.1874360740184784,
|
| 62 |
"learning_rate": 6.307399704769099e-05,
|
| 63 |
+
"loss": 1.093,
|
| 64 |
"step": 90
|
| 65 |
},
|
| 66 |
{
|
| 67 |
+
"grad_norm": 0.20323705673217773,
|
| 68 |
"learning_rate": 5.495227651252315e-05,
|
| 69 |
+
"loss": 1.0979,
|
| 70 |
"step": 100
|
| 71 |
},
|
| 72 |
{
|
| 73 |
+
"grad_norm": 0.2660678029060364,
|
| 74 |
"learning_rate": 4.669547078371504e-05,
|
| 75 |
+
"loss": 1.0988,
|
| 76 |
"step": 110
|
| 77 |
},
|
| 78 |
{
|
| 79 |
+
"grad_norm": 0.18318524956703186,
|
| 80 |
"learning_rate": 3.852880399766243e-05,
|
| 81 |
+
"loss": 1.0875,
|
| 82 |
"step": 120
|
| 83 |
},
|
| 84 |
{
|
| 85 |
+
"grad_norm": 0.16894328594207764,
|
| 86 |
"learning_rate": 3.0675041535377405e-05,
|
| 87 |
+
"loss": 1.0838,
|
| 88 |
"step": 130
|
| 89 |
},
|
| 90 |
{
|
| 91 |
+
"grad_norm": 0.28183799982070923,
|
| 92 |
"learning_rate": 2.3348413563600325e-05,
|
| 93 |
+
"loss": 1.0836,
|
| 94 |
"step": 140
|
| 95 |
},
|
| 96 |
{
|
| 97 |
+
"grad_norm": 0.1599818468093872,
|
| 98 |
"learning_rate": 1.6748771394307585e-05,
|
| 99 |
+
"loss": 1.0826,
|
| 100 |
"step": 150
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"grad_norm": 0.19995185732841492,
|
| 104 |
"learning_rate": 1.1056136061894384e-05,
|
| 105 |
+
"loss": 1.0854,
|
| 106 |
"step": 160
|
| 107 |
},
|
| 108 |
{
|
| 109 |
+
"grad_norm": 0.1766791194677353,
|
| 110 |
"learning_rate": 6.425787818636131e-06,
|
| 111 |
+
"loss": 1.0848,
|
| 112 |
"step": 170
|
| 113 |
},
|
| 114 |
{
|
| 115 |
+
"grad_norm": 0.1526567041873932,
|
| 116 |
"learning_rate": 2.9840304941919415e-06,
|
| 117 |
+
"loss": 1.0879,
|
| 118 |
"step": 180
|
| 119 |
},
|
| 120 |
{
|
| 121 |
+
"grad_norm": 0.18048691749572754,
|
| 122 |
"learning_rate": 8.247462563808817e-07,
|
| 123 |
+
"loss": 1.0787,
|
| 124 |
"step": 190
|
| 125 |
},
|
| 126 |
{
|
| 127 |
+
"grad_norm": 0.17072460055351257,
|
| 128 |
"learning_rate": 6.834750376549792e-09,
|
| 129 |
+
"loss": 1.0785,
|
| 130 |
"step": 200
|
| 131 |
}
|
| 132 |
],
|
|
|
|
| 148 |
}
|
| 149 |
},
|
| 150 |
"total_flos": 0.0,
|
| 151 |
+
"train_batch_size": 8,
|
| 152 |
"trial_name": null,
|
| 153 |
"trial_params": null
|
| 154 |
}
|
checkpoint-200/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 7825
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8948bc0acba8971f43e26c05e378b8f61c5f795118c5779a55a4b9bba2bbeb33
|
| 3 |
size 7825
|