Upload folder using huggingface_hub

.gitattributes

@@ -380,3 +380,8 @@ sft/665K36/revise_Full_smoe_sharev3/analysts/0717_2301_llava...mstar_llava_model
 sft/665K36/revise_Full_smoe_sharev3/analysts/0717_2301_llava...mstar_llava_model_args_fe4e53/mme.json filter=lfs diff=lfs merge=lfs -text
 sft/665K36/revise_Full_smoe_sharev3/analysts/0717_2301_llava...mstar_llava_model_args_fe4e53/mmmu_val.json filter=lfs diff=lfs merge=lfs -text
 sft/665K36/revise_Full_smoe_sharev3/analysts/0717_2301_llava...mstar_llava_model_args_fe4e53/mmstar.json filter=lfs diff=lfs merge=lfs -text
+sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/infovqa_val.json filter=lfs diff=lfs merge=lfs -text
+sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/mmbench_en_dev.json filter=lfs diff=lfs merge=lfs -text
+sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/mmerealworld_lite.json filter=lfs diff=lfs merge=lfs -text
+sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/submissions/mmbench_en_dev_results.xlsx filter=lfs diff=lfs merge=lfs -text
+sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/textvqa_val.json filter=lfs diff=lfs merge=lfs -text

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/infovqa_val.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29a7d0a6b377e2db41a41138f9c0165be53761898d5871f29d3ba09e749d3082
+size 576426884

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/mmbench_en_dev.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1160459a8adad31add5915f0c989f40dd9d2a73dedeb89365eed11ddac2b0af
+size 14568288

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/mmerealworld_lite.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d071cdf852b01c5d7548b8130aada4f8e15fda5c595a5a840ddb48d3b6bd24ae
+size 1994104321

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/rank0_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 0 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/rank1_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 1 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/rank2_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 2 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/rank3_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 3 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/results.json

@@ -0,0 +1,729 @@
+{
+    "results": {
+        "ai2d": {
+            "exact_match,flexible-extract": 0.645401554404145,
+            "exact_match_stderr,flexible-extract": 0.008610236936498226,
+            "alias": "ai2d"
+        },
+        "infovqa_val": {
+            "anls,none": 0.26811852909675116,
+            "anls_stderr,none": 0.008371528778539056,
+            "alias": "infovqa_val"
+        },
+        "mmbench_en_dev": {
+            "gpt_eval_score,none": 72.42268041237114,
+            "gpt_eval_score_stderr,none": "N/A",
+            "submission,none": null,
+            "submission_stderr,none": "N/A",
+            "alias": "mmbench_en_dev"
+        },
+        "mme": {
+            "mme_cognition_score,none": 305.3571428571429,
+            "mme_cognition_score_stderr,none": "N/A",
+            "mme_percetion_score,none": 1400.8141256502602,
+            "mme_percetion_score_stderr,none": "N/A",
+            "alias": "mme"
+        },
+        "mmerealworld_lite": {
+            "mme_realworld_score,none": 0.31474726420010424,
+            "mme_realworld_score_stderr,none": "N/A",
+            "alias": "mmerealworld_lite"
+        },
+        "mmmu_pro_standard": {
+            "mmmu_acc,none": 0.25838,
+            "mmmu_acc_stderr,none": "N/A",
+            "alias": "mmmu_pro_standard"
+        },
+        "mmmu_val": {
+            "mmmu_acc,none": 0.42444,
+            "mmmu_acc_stderr,none": "N/A",
+            "alias": "mmmu_val"
+        },
+        "mmstar": {
+            "coarse perception,none": 0.7007139127935286,
+            "coarse perception_stderr,none": "N/A",
+            "fine-grained perception,none": 0.36133628101203635,
+            "fine-grained perception_stderr,none": "N/A",
+            "instance reasoning,none": 0.5332709723793451,
+            "instance reasoning_stderr,none": "N/A",
+            "logical reasoning,none": 0.37952487556447956,
+            "logical reasoning_stderr,none": "N/A",
+            "math,none": 0.2904916243428673,
+            "math_stderr,none": "N/A",
+            "science & technology,none": 0.2396233390154238,
+            "science & technology_stderr,none": "N/A",
+            "alias": "mmstar"
+        },
+        "scienceqa_img": {
+            "exact_match,none": 0.7481408031730292,
+            "exact_match_stderr,none": 0.009667747830565338,
+            "alias": "scienceqa_img"
+        },
+        "seedbench_2_plus": {
+            "seedbench_2_plus_Chart,none": 0.4765432098765432,
+            "seedbench_2_plus_Chart_stderr,none": "N/A",
+            "seedbench_2_plus_all,none": 0.47474747474747475,
+            "seedbench_2_plus_all_stderr,none": "N/A",
+            "seedbench_2_plus_Web,none": 0.509090909090909,
+            "seedbench_2_plus_Web_stderr,none": "N/A",
+            "seedbench_2_plus_Map,none": 0.4448574969021066,
+            "seedbench_2_plus_Map_stderr,none": "N/A",
+            "alias": "seedbench_2_plus"
+        },
+        "textvqa_val": {
+            "exact_match,none": 0.4206200003147125,
+            "exact_match_stderr,none": 0.0067633506191732045,
+            "submission,none": null,
+            "submission_stderr,none": "N/A",
+            "alias": "textvqa_val"
+        }
+    },
+    "configs": {
+        "ai2d": {
+            "task": "ai2d",
+            "dataset_path": "lmms-lab/ai2d",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function ai2d_doc_to_visual at 0x7f5391f408b0>",
+            "doc_to_text": "<function ai2d_doc_to_text at 0x7f5391f40b80>",
+            "doc_to_target": "<function ai2d_doc_to_target at 0x7f5391f48160>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "exact_match",
+                    "aggregation": "mean",
+                    "higher_is_better": true,
+                    "ignore_case": true,
+                    "ignore_punctuation": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 512,
+                "temperature": 0.0,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "filter_list": [
+                {
+                    "name": "flexible-extract",
+                    "filter": [
+                        {
+                            "function": "<class 'utils.MultiChoiceRegexFilter'>",
+                            "group_select": 0,
+                            "ignore_case": true,
+                            "ignore_punctuation": true,
+                            "regex_pattern": "([A-Z])\\."
+                        }
+                    ]
+                }
+            ],
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "prompt_format": "mcq",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer with the option's letter from the given choices directly."
+                },
+                "gpt4v": {
+                    "prompt_format": "mcq",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAbove choices are given in {option}. {content} format.\nPlease answer with the option letter from the given choices directly."
+                },
+                "qwen_vl": {
+                    "prompt_format": "qa",
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                },
+                "xcomposer2_4khd": {
+                    "prompt_format": "mcq_xcomposer",
+                    "pre_prompt": "[UNUSED_TOKEN_146]user\nQuestion: ",
+                    "post_prompt": "[UNUSED_TOKEN_145]\n[UNUSED_TOKEN_146]assistant\nThe answer is"
+                }
+            },
+            "model_specific_target_kwargs": {
+                "default": "mcq",
+                "qwen_vl": "qa"
+            }
+        },
+        "infovqa_val": {
+            "task": "infovqa_val",
+            "dataset_path": "lmms-lab/DocVQA",
+            "dataset_name": "InfographicVQA",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "validation",
+            "doc_to_visual": "<function infovqa_doc_to_visual at 0x7f53c4b55940>",
+            "doc_to_text": "<function infovqa_doc_to_text at 0x7f53c4b55c10>",
+            "doc_to_target": "answers",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "anls",
+                    "aggregation": "mean",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 32,
+                "temperature": 0.0,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer the question using a single word or phrase."
+                }
+            }
+        },
+        "mmbench_en_dev": {
+            "task": "mmbench_en_dev",
+            "dataset_path": "lmms-lab/MMBench",
+            "dataset_name": "en",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "dev",
+            "doc_to_visual": "<function mmbench_doc_to_visual at 0x7f53f11925e0>",
+            "doc_to_text": "<function mmbench_doc_to_text at 0x7f53f1192af0>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmbench_process_results at 0x7f53f119e040>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "gpt_eval_score",
+                    "aggregation": "<function mmbench_aggregate_dev_results_eval at 0x7f53f1188e50>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "submission",
+                    "aggregation": "<function mmbench_aggregate_dev_results_submission at 0x7f53f11923a0>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "until": [
+                    "ASSISTANT:"
+                ],
+                "max_new_tokens": 1024,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "image_aspect_ratio": "original"
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer with the option's letter from the given choices directly."
+                }
+            },
+            "model_specific_generation_kwargs": {
+                "llava": {
+                    "image_aspect_ratio": "original"
+                }
+            }
+        },
+        "mme": {
+            "task": "mme",
+            "dataset_path": "lmms-lab/MME",
+            "dataset_kwargs": {
+                "token": false
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function mme_doc_to_visual at 0x7f53f2c6baf0>",
+            "doc_to_text": "<function mme_doc_to_text at 0x7f53f1213310>",
+            "doc_to_target": "answer",
+            "process_results": "<function mme_process_results at 0x7f53f12138b0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mme_percetion_score",
+                    "aggregation": "<function mme_aggregate_results at 0x7f53f1213dc0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "mme_cognition_score",
+                    "aggregation": "<function mme_aggregate_results at 0x7f53f121a280>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 16,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer the question using a single word or phrase."
+                },
+                "gpt4v": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer the question with Yes or No."
+                },
+                "qwen_vl": {
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                },
+                "otterhd": {
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                },
+                "xcomposer2_4khd": {
+                    "pre_prompt": "[UNUSED_TOKEN_146]user\n",
+                    "post_prompt": " Answer this question briefly[UNUSED_TOKEN_145]\n[UNUSED_TOKEN_146]assistant\n"
+                }
+            }
+        },
+        "mmerealworld_lite": {
+            "task": "mmerealworld_lite",
+            "dataset_path": "yifanzhang114/MME-RealWorld-lite-lmms-eval",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "train",
+            "doc_to_visual": "<function mme_realworld_doc_to_visual at 0x7f53926e9a60>",
+            "doc_to_text": "<function mme_realworld_doc_to_text at 0x7f53926ed550>",
+            "doc_to_target": "answer",
+            "process_results": "<function mme_realworld_process_results at 0x7f53926f31f0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mme_realworld_score",
+                    "aggregation": "<function mme_realworld_aggregate_results at 0x7f53926f3d30>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 16,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nSelect the best answer to the above multiple-choice question based on the image. Respond with only the letter (A, B, C, D, or E) of the correct option."
+                },
+                "gpt4v": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nSelect the best answer to the above multiple-choice question based on the image. Respond with only the letter (A, B, C, D, or E) of the correct option."
+                },
+                "xcomposer2_4khd": {
+                    "pre_prompt": "[UNUSED_TOKEN_146]user\n",
+                    "post_prompt": " Answer this question with A, B, C, or D.[UNUSED_TOKEN_145]\n[UNUSED_TOKEN_146]assistant\n"
+                }
+            }
+        },
+        "mmmu_pro_standard": {
+            "task": "mmmu_pro_standard",
+            "dataset_path": "MMMU/MMMU_Pro",
+            "dataset_name": "standard (10 options)",
+            "test_split": "test",
+            "doc_to_visual": "<function mmmu_pro_doc_to_visual at 0x7f53f115f940>",
+            "doc_to_text": "<function mmmu_pro_doc_to_text at 0x7f53f116c700>",
+            "doc_to_target": "{{answer}}",
+            "process_results": "<function mmmu_pro_process_results at 0x7f53f1172670>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mmmu_acc",
+                    "aggregation": "<function mmmu_pro_aggregate_results at 0x7f53f117e5e0>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 256,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": {
+                "version": 0.0,
+                "interleaved_format": false
+            },
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "Answer with the option letter from the given choices directly."
+                }
+            }
+        },
+        "mmmu_val": {
+            "task": "mmmu_val",
+            "dataset_path": "lmms-lab/MMMU",
+            "test_split": "validation",
+            "doc_to_visual": "<function mmmu_doc_to_visual at 0x7f53a412b8b0>",
+            "doc_to_text": "<function mmmu_doc_to_text at 0x7f53a40a95e0>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmmu_process_results at 0x7f53a40764c0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mmmu_acc",
+                    "aggregation": "<function mmmu_aggregate_results at 0x7f53a3eaf430>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 128,
+                "until": [
+                    "\n\n"
+                ],
+                "image_aspect_ratio": "original"
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_generation_kwargs": {
+                "llava": {
+                    "image_aspect_ratio": "original"
+                }
+            }
+        },
+        "mmstar": {
+            "task": "mmstar",
+            "dataset_path": "Lin-Chen/MMStar",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "val",
+            "doc_to_visual": "<function mmstar_doc_to_visual at 0x7f5391e97ee0>",
+            "doc_to_text": "<function mmstar_doc_to_text at 0x7f5391ea33a0>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmstar_process_results at 0x7f5391ea38b0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "coarse perception",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f5391ea3d30>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "fine-grained perception",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f5391e27160>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "instance reasoning",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f5391e27550>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "logical reasoning",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f5391e27940>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "science & technology",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f5391e27d30>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "math",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f5391e2e160>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "until": [
+                    "\n\n"
+                ],
+                "do_sample": false
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer with the option's letter from the given choices directly"
+                }
+            }
+        },
+        "scienceqa_img": {
+            "task": "scienceqa_img",
+            "dataset_path": "lmms-lab/ScienceQA",
+            "dataset_name": "ScienceQA-IMG",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function sqa_doc_to_visual at 0x7f53bcffdf70>",
+            "doc_to_text": "<function sqa_doc_to_text at 0x7f53bceb71f0>",
+            "doc_to_target": "<function sqa_doc_to_target at 0x7f53bceb75e0>",
+            "process_results": "<function sqa_process_results at 0x7f53bceb7940>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "exact_match",
+                    "aggregation": "mean",
+                    "higher_is_better": true,
+                    "ignore_case": true,
+                    "ignore_punctuation": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 16,
+                "temperature": 0.0,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ],
+                "image_aspect_ratio": "original"
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "format": "default",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer with the option's letter from the given choices directly."
+                },
+                "qwen_vl": {
+                    "format": "qwen_vl"
+                },
+                "idefics2": {
+                    "format": "default",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer:"
+                }
+            },
+            "model_specific_generation_kwargs": {
+                "llava": {
+                    "image_aspect_ratio": "original"
+                }
+            }
+        },
+        "seedbench_2_plus": {
+            "task": "seedbench_2_plus",
+            "dataset_path": "doolayer/SEED-Bench-2-Plus",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function seed_doc_to_visual at 0x7f53b9d8a4c0>",
+            "doc_to_text": "<function seed_doc_to_text at 0x7f53b9d8aaf0>",
+            "doc_to_target": "answer",
+            "process_results": "<function seed_process_result at 0x7f53b9ab5040>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "seedbench_2_plus_Chart",
+                    "aggregation": "<function seed_aggregation_result at 0x7f53b9ab5550>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "seedbench_2_plus_Map",
+                    "aggregation": "<function seed_aggregation_result at 0x7f53b9ab59d0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "seedbench_2_plus_Web",
+                    "aggregation": "<function seed_aggregation_result at 0x7f53b9ab5e50>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "seedbench_2_plus_all",
+                    "aggregation": "<function seed_aggregation_result at 0x7f53b93d5310>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "until": [
+                    "ASSISTANT:"
+                ],
+                "max_new_tokens": 16,
+                "image_aspect_ratio": "original"
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "llava": {
+                    "img_token": "<image>",
+                    "post_prompt": "Answer with the option's letter from the given choices directly."
+                },
+                "gpt4V": {
+                    "img_token": "<image>",
+                    "post_prompt": "Answer with the option's letter from the given choices directly."
+                },
+                "default": {
+                    "img_token": "<image>",
+                    "post_prompt": "Answer with the option's letter from the given choices directly."
+                }
+            }
+        },
+        "textvqa_val": {
+            "task": "textvqa_val",
+            "dataset_path": "lmms-lab/textvqa",
+            "test_split": "validation",
+            "doc_to_visual": "<function textvqa_doc_to_visual at 0x7f53be318b80>",
+            "doc_to_text": "<function textvqa_doc_to_text at 0x7f53be333040>",
+            "doc_to_target": "answer",
+            "process_results": "<function textvqa_process_results at 0x7f53be333310>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "exact_match",
+                    "aggregation": "mean",
+                    "higher_is_better": true,
+                    "ignore_case": true,
+                    "ignore_punctuation": true
+                },
+                {
+                    "metric": "submission",
+                    "aggregation": "<function textvqa_aggregate_submissions at 0x7f53be318940>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "until": [
+                    "ASSISTANT:"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer the question using a single word or phrase.",
+                    "ocr": false
+                },
+                "qwen_vl": {
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                }
+            }
+        }
+    },
+    "versions": {
+        "ai2d": "Yaml",
+        "infovqa_val": "Yaml",
+        "mmbench_en_dev": "Yaml",
+        "mme": "Yaml",
+        "mmerealworld_lite": "Yaml",
+        "mmmu_pro_standard": "Yaml",
+        "mmmu_val": "Yaml",
+        "mmstar": "Yaml",
+        "scienceqa_img": "Yaml",
+        "seedbench_2_plus": "Yaml",
+        "textvqa_val": "Yaml"
+    },
+    "n-shot": {
+        "ai2d": 0,
+        "infovqa_val": 0,
+        "mmbench_en_dev": 0,
+        "mme": 0,
+        "mmerealworld_lite": 0,
+        "mmmu_pro_standard": 0,
+        "mmmu_val": 0,
+        "mmstar": 0,
+        "scienceqa_img": 0,
+        "seedbench_2_plus": 0,
+        "textvqa_val": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78_new/revise_checkpoints/Xphi35-siglip224/SMOE/665K36/revise_Full_smoe_sharev3,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/submissions/mmbench_en_dev_results.json

@@ -0,0 +1 @@
+{"overall_acc": 0.7242268041237113, "category_acc": {"action_recognition": 0.8888888888888888, "attribute_comparison": 0.7272727272727273, "attribute_recognition": 0.8648648648648649, "celebrity_recognition": 0.7878787878787878, "function_reasoning": 0.8607594936708861, "future_prediction": 0.525, "identity_reasoning": 0.9777777777777777, "image_emotion": 0.82, "image_quality": 0.5094339622641509, "image_scene": 0.9615384615384616, "image_style": 0.9056603773584906, "image_topic": 0.8611111111111112, "nature_relation": 0.6041666666666666, "object_localization": 0.5308641975308642, "ocr": 0.717948717948718, "physical_property_reasoning": 0.5866666666666667, "physical_relation": 0.4583333333333333, "social_relation": 0.7906976744186046, "spatial_relationship": 0.3333333333333333, "structuralized_imagetext_understanding": 0.47435897435897434}, "l2_category_acc": {"attribute_reasoning": 0.7839195979899497, "coarse_perception": 0.8344594594594594, "finegrained_perception (cross-instance)": 0.6643356643356644, "finegrained_perception (instance-level)": 0.726962457337884, "logic_reasoning": 0.4915254237288136, "relation_reasoning": 0.6434782608695652}}

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/submissions/mmbench_en_dev_results.xlsx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c865ae8c4bebda3a05bc5d1d735de59e432ba4ef891218b9b6aba6f23fd33d8f
+size 865128

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1127_llava...l_mme_llava_model_args_5a2714/textvqa_val.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2008f45cb5e06d281e3c245fa4123d5cb8f5f42dbbceecd0c544301fea7fc47
+size 13146759

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1201_llava...image_llava_model_args_5a2714/rank0_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 0 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1201_llava...image_llava_model_args_5a2714/rank1_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 1 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1201_llava...image_llava_model_args_5a2714/rank2_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 2 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1201_llava...image_llava_model_args_5a2714/rank3_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 3 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1201_llava...image_llava_model_args_5a2714/results.json

@@ -0,0 +1,146 @@
+{
+    "results": {
+        "hallusion_bench_image": {
+            "aAcc,none": 41.4301,
+            "aAcc_stderr,none": "N/A",
+            "fAcc,none": 17.052,
+            "fAcc_stderr,none": "N/A",
+            "qAcc,none": 11.4286,
+            "qAcc_stderr,none": "N/A",
+            "alias": "hallusion_bench_image"
+        },
+        "mathvista_testmini": {
+            "gpt_eval_score,none": 30.6,
+            "gpt_eval_score_stderr,none": "N/A",
+            "alias": "mathvista_testmini"
+        }
+    },
+    "configs": {
+        "hallusion_bench_image": {
+            "task": "hallusion_bench_image",
+            "dataset_path": "lmms-lab/HallusionBench",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "image",
+            "doc_to_visual": "<function hb_doc_to_visual at 0x7f66394cf4c0>",
+            "doc_to_text": "<function hb_doc_to_text at 0x7f66394cfc10>",
+            "doc_to_target": "gt_answer_details",
+            "process_results": "<function hb_process_results at 0x7f6639462430>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "aAcc",
+                    "aggregation": "<function hb_aggregation_result_aAcc at 0x7f6639462dc0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "qAcc",
+                    "aggregation": "<function hb_aggregation_result_qAcc at 0x7f663941b430>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "fAcc",
+                    "aggregation": "<function hb_aggregation_result_fAcc at 0x7f663941bc10>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 128,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": ""
+                }
+            }
+        },
+        "mathvista_testmini": {
+            "task": "mathvista_testmini",
+            "dataset_path": "AI4Math/MathVista",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "testmini",
+            "doc_to_visual": "<function mathvista_doc_to_visual at 0x7f660f16a8b0>",
+            "doc_to_text": "<function mathvista_doc_to_text at 0x7f660f172f70>",
+            "doc_to_target": "answer",
+            "process_results": "<function mathvista_process_results at 0x7f660f180670>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "gpt_eval_score",
+                    "aggregation": "<function mathvista_aggregate_results at 0x7f660f18bd30>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "until": [
+                    "ASSISTANT:"
+                ],
+                "max_new_tokens": 1024,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "image_aspect_ratio": "original"
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "shot_type": "format-prompt",
+                    "shot": 0,
+                    "use_caption": false,
+                    "use_ocr": false
+                },
+                "phi3v": {
+                    "shot_type": "solution"
+                }
+            },
+            "model_specific_generation_kwargs": {
+                "llava": {
+                    "image_aspect_ratio": "original"
+                }
+            }
+        }
+    },
+    "versions": {
+        "hallusion_bench_image": "Yaml",
+        "mathvista_testmini": "Yaml"
+    },
+    "n-shot": {
+        "hallusion_bench_image": 0,
+        "mathvista_testmini": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78_new/revise_checkpoints/Xphi35-siglip224/SMOE/665K36/revise_Full_smoe_sharev3,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1220_llava...bench_llava_model_args_5a2714/rank0_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 0 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1220_llava...bench_llava_model_args_5a2714/rank1_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 1 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1220_llava...bench_llava_model_args_5a2714/rank2_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 2 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1220_llava...bench_llava_model_args_5a2714/rank3_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 3 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1220_llava...bench_llava_model_args_5a2714/results.json

@@ -0,0 +1,67 @@
+{
+    "results": {
+        "ocrbench": {
+            "ocrbench_accuracy,none": 0.326,
+            "ocrbench_accuracy_stderr,none": "N/A",
+            "alias": "ocrbench"
+        }
+    },
+    "configs": {
+        "ocrbench": {
+            "task": "ocrbench",
+            "dataset_path": "echo840/OCRBench",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function ocrbench_doc_to_visual at 0x7fd6ac655f70>",
+            "doc_to_text": "<function ocrbench_doc_to_text at 0x7fd69a22e550>",
+            "doc_to_target": "answer",
+            "process_results": "<function ocrbench_process_results at 0x7fd69a22e8b0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "ocrbench_accuracy",
+                    "aggregation": "<function ocrbench_aggregate_accuracy at 0x7fd69a22ec10>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 128,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ]
+        }
+    },
+    "versions": {
+        "ocrbench": "Yaml"
+    },
+    "n-shot": {
+        "ocrbench": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78_new/revise_checkpoints/Xphi35-siglip224/SMOE/665K36/revise_Full_smoe_sharev3,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1220_llava...bench_llava_model_args_5a2714/results/ocrbench_results.txt

@@ -0,0 +1,18 @@
+######################### OCRBench #############################
+Text Recognition(Total 300): 183
+---------------- Details of Recognition Score ------------------
+Regular Text Recognition(Total 50): 45
+Irregular Text Recognition(Total 50): 43
+Artistic Text Recognition(Total 50): 46
+Handwriting Recognition(Total 50): 28
+Digit String Recognition(Total 50): 13
+Non-Semantic Text Recognition(Total 50): 8
+----------------------------------------------------------------
+Scene Text-centric VQA(Total 200): 114
+----------------------------------------------------------------
+Doc-oriented VQA(Total 200): 20
+----------------------------------------------------------------
+Key Information Extraction(Total 200): 9
+Handwritten Mathematical Expression Recognition(Total 100): 0
+--------------------- Final Score ------------------------------
+Final Score(Total 1000): 326

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1739_llava_v1.5_ai2d_llava_model_args_5a2714/rank0_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 0 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1739_llava_v1.5_ai2d_llava_model_args_5a2714/rank1_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 1 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1739_llava_v1.5_ai2d_llava_model_args_5a2714/rank2_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 2 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1739_llava_v1.5_ai2d_llava_model_args_5a2714/rank3_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 3 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1739_llava_v1.5_ai2d_llava_model_args_5a2714/results.json

@@ -0,0 +1,106 @@
+{
+    "results": {
+        "ai2d": {
+            "exact_match,flexible-extract": 0.6463730569948186,
+            "exact_match_stderr,flexible-extract": 0.008604903043803526,
+            "alias": "ai2d"
+        }
+    },
+    "configs": {
+        "ai2d": {
+            "task": "ai2d",
+            "dataset_path": "lmms-lab/ai2d",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function ai2d_doc_to_visual at 0x7f74c14778b0>",
+            "doc_to_text": "<function ai2d_doc_to_text at 0x7f74c1477b80>",
+            "doc_to_target": "<function ai2d_doc_to_target at 0x7f74c147f160>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "exact_match",
+                    "aggregation": "mean",
+                    "higher_is_better": true,
+                    "ignore_case": true,
+                    "ignore_punctuation": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 512,
+                "temperature": 0.0,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "filter_list": [
+                {
+                    "name": "flexible-extract",
+                    "filter": [
+                        {
+                            "function": "<class 'utils.MultiChoiceRegexFilter'>",
+                            "group_select": 0,
+                            "ignore_case": true,
+                            "ignore_punctuation": true,
+                            "regex_pattern": "([A-Z])\\."
+                        }
+                    ]
+                }
+            ],
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "prompt_format": "mcq",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer with the option's letter from the given choices directly."
+                },
+                "gpt4v": {
+                    "prompt_format": "mcq",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAbove choices are given in {option}. {content} format.\nPlease answer with the option letter from the given choices directly."
+                },
+                "qwen_vl": {
+                    "prompt_format": "qa",
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                },
+                "xcomposer2_4khd": {
+                    "prompt_format": "mcq_xcomposer",
+                    "pre_prompt": "[UNUSED_TOKEN_146]user\nQuestion: ",
+                    "post_prompt": "[UNUSED_TOKEN_145]\n[UNUSED_TOKEN_146]assistant\nThe answer is"
+                }
+            },
+            "model_specific_target_kwargs": {
+                "default": "mcq",
+                "qwen_vl": "qa"
+            }
+        }
+    },
+    "versions": {
+        "ai2d": "Yaml"
+    },
+    "n-shot": {
+        "ai2d": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78_new/revise_checkpoints/Xphi35-siglip224/SMOE/665K36/revise_Full_smoe_sharev3,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1743_llava_v1.5_ai2d_llava_model_args_5a2714/rank0_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 0 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1743_llava_v1.5_ai2d_llava_model_args_5a2714/rank1_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 1 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1743_llava_v1.5_ai2d_llava_model_args_5a2714/rank2_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 2 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1743_llava_v1.5_ai2d_llava_model_args_5a2714/rank3_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 3 eval done

sft/665K36/revise_Full_smoe_sharev3/logs/0717_1743_llava_v1.5_ai2d_llava_model_args_5a2714/results.json

@@ -0,0 +1,106 @@
+{
+    "results": {
+        "ai2d": {
+            "exact_match,flexible-extract": 0.6541450777202072,
+            "exact_match_stderr,flexible-extract": 0.008560826682457891,
+            "alias": "ai2d"
+        }
+    },
+    "configs": {
+        "ai2d": {
+            "task": "ai2d",
+            "dataset_path": "lmms-lab/ai2d",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function ai2d_doc_to_visual at 0x7f1b188ff940>",
+            "doc_to_text": "<function ai2d_doc_to_text at 0x7f1b188ffc10>",
+            "doc_to_target": "<function ai2d_doc_to_target at 0x7f1b189061f0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "exact_match",
+                    "aggregation": "mean",
+                    "higher_is_better": true,
+                    "ignore_case": true,
+                    "ignore_punctuation": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 512,
+                "temperature": 0.0,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "filter_list": [
+                {
+                    "name": "flexible-extract",
+                    "filter": [
+                        {
+                            "function": "<class 'utils.MultiChoiceRegexFilter'>",
+                            "group_select": 0,
+                            "ignore_case": true,
+                            "ignore_punctuation": true,
+                            "regex_pattern": "([A-Z])\\."
+                        }
+                    ]
+                }
+            ],
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "prompt_format": "mcq",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer with the option's letter from the given choices directly."
+                },
+                "gpt4v": {
+                    "prompt_format": "mcq",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAbove choices are given in {option}. {content} format.\nPlease answer with the option letter from the given choices directly."
+                },
+                "qwen_vl": {
+                    "prompt_format": "qa",
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                },
+                "xcomposer2_4khd": {
+                    "prompt_format": "mcq_xcomposer",
+                    "pre_prompt": "[UNUSED_TOKEN_146]user\nQuestion: ",
+                    "post_prompt": "[UNUSED_TOKEN_145]\n[UNUSED_TOKEN_146]assistant\nThe answer is"
+                }
+            },
+            "model_specific_target_kwargs": {
+                "default": "mcq",
+                "qwen_vl": "qa"
+            }
+        }
+    },
+    "versions": {
+        "ai2d": "Yaml"
+    },
+    "n-shot": {
+        "ai2d": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78_new/revise_checkpoints/Xphi35-siglip224/SMOE/665K36/revise_Full_smoe_sharev3,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}