Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,41 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs1/daru_treewayextractive.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs1/darumeru_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs1/darumeru_ruTiE.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs1/nlpcoreteam_enMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs1/nlpcoreteam_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs16/daru_treewayextractive.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs16/darumeru_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs16/nlpcoreteam_enMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs16/nlpcoreteam_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs4/daru_treewayextractive.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs4/darumeru_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs4/darumeru_ruTiE.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs4/nlpcoreteam_enMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k0_bs4/nlpcoreteam_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs1/daru_treewayabstractive.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs1/daru_treewayextractive.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs1/darumeru_MultiQ.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs1/darumeru_USE.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs1/darumeru_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs1/darumeru_ruOpenBookQA.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs1/darumeru_ruTiE.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs1/nlpcoreteam_enMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs1/nlpcoreteam_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs1/russiannlp_rucola_custom.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs16/daru_treewayextractive.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs16/darumeru_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs16/nlpcoreteam_enMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs16/nlpcoreteam_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs4/daru_treewayabstractive.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs4/daru_treewayextractive.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs4/darumeru_MultiQ.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs4/darumeru_USE.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs4/darumeru_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs4/darumeru_ruOpenBookQA.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs4/darumeru_ruTiE.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs4/nlpcoreteam_enMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs4/nlpcoreteam_ruMMLU.jsonl filter=lfs diff=lfs merge=lfs -text
+llmtf_eval_k5_bs4/russiannlp_rucola_custom.jsonl filter=lfs diff=lfs merge=lfs -text

llmtf_eval_k0_bs1/daru_treewayabstractive_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 512,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 500,
+        "method": "generate"
+    }
+}

llmtf_eval_k0_bs1/daru_treewayabstractive_total.jsonl

@@ -0,0 +1,8 @@
+{
+    "task_name": "daru/treewayabstractive",
+    "results": {
+        "rouge1": 0.35429775115253526,
+        "rouge2": 0.12889173695955833
+    },
+    "leaderboard_result": 0.2415947440560468
+}

llmtf_eval_k0_bs1/daru_treewayextractive.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ffa3ec35722ce316fdaa3af761f8797f6db0f94a32596f1e05741e81068d3d9f
+size 259989896

llmtf_eval_k0_bs1/daru_treewayextractive_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 1,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 500,
+        "method": "calculate_logsoftmax"
+    }
+}

llmtf_eval_k0_bs1/daru_treewayextractive_total.jsonl

@@ -0,0 +1,7 @@
+{
+    "task_name": "daru/treewayextractive",
+    "results": {
+        "r-prec": 0.3960751082251082
+    },
+    "leaderboard_result": 0.3960751082251082
+}

llmtf_eval_k0_bs1/darumeru_MultiQ_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 64,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "generate"
+    }
+}

llmtf_eval_k0_bs1/darumeru_MultiQ_total.jsonl

@@ -0,0 +1,8 @@
+{
+    "task_name": "darumeru/MultiQ",
+    "results": {
+        "f1": 0.33947318357368605,
+        "em": 0.21414913957934992
+    },
+    "leaderboard_result": 0.276811161576518
+}

llmtf_eval_k0_bs1/darumeru_PARus_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 64,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "calculate_tokens_proba"
+    }
+}

llmtf_eval_k0_bs1/darumeru_PARus_total.jsonl

@@ -0,0 +1,7 @@
+{
+    "task_name": "darumeru/PARus",
+    "results": {
+        "acc": 0.66
+    },
+    "leaderboard_result": 0.66
+}

llmtf_eval_k0_bs1/darumeru_RCB_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 64,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "calculate_tokens_proba"
+    }
+}

llmtf_eval_k0_bs1/darumeru_RCB_total.jsonl

@@ -0,0 +1,8 @@
+{
+    "task_name": "darumeru/RCB",
+    "results": {
+        "acc": 0.5,
+        "f1_macro": 0.43018975381906
+    },
+    "leaderboard_result": 0.46509487690953
+}

llmtf_eval_k0_bs1/darumeru_RWSD_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 64,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "calculate_tokens_proba"
+    }
+}

llmtf_eval_k0_bs1/darumeru_RWSD_total.jsonl

@@ -0,0 +1,7 @@
+{
+    "task_name": "darumeru/RWSD",
+    "results": {
+        "acc": 0.5441176470588235
+    },
+    "leaderboard_result": 0.5441176470588235
+}

llmtf_eval_k0_bs1/darumeru_USE_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 64,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "generate"
+    }
+}

llmtf_eval_k0_bs1/darumeru_USE_total.jsonl

@@ -0,0 +1,7 @@
+{
+    "task_name": "darumeru/USE",
+    "results": {
+        "grade_norm": 0.07941176470588233
+    },
+    "leaderboard_result": 0.07941176470588233
+}

llmtf_eval_k0_bs1/darumeru_cp_para_en_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 1024,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "generate"
+    }
+}

llmtf_eval_k0_bs1/darumeru_cp_para_en_total.jsonl

@@ -0,0 +1,9 @@
+{
+    "task_name": "darumeru/cp_para_en",
+    "results": {
+        "symbol_per_token": 4.463061170262149,
+        "len": 0.9941296296409974,
+        "lcs": 0.9527227031116661
+    },
+    "leaderboard_result": 0.9527227031116661
+}

llmtf_eval_k0_bs1/darumeru_cp_para_ru_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 1024,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "generate"
+    }
+}

llmtf_eval_k0_bs1/darumeru_cp_para_ru_total.jsonl

@@ -0,0 +1,9 @@
+{
+    "task_name": "darumeru/cp_para_ru",
+    "results": {
+        "symbol_per_token": 2.965959526577639,
+        "len": 0.9943389906706583,
+        "lcs": 0.9089360152526831
+    },
+    "leaderboard_result": 0.9089360152526831
+}

llmtf_eval_k0_bs1/darumeru_cp_sent_en_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 128,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "generate"
+    }
+}

llmtf_eval_k0_bs1/darumeru_cp_sent_en_total.jsonl

@@ -0,0 +1,9 @@
+{
+    "task_name": "darumeru/cp_sent_en",
+    "results": {
+        "symbol_per_token": 4.424738337449315,
+        "len": 0.9996416196590585,
+        "lcs": 0.9958136839407484
+    },
+    "leaderboard_result": 0.9996416196590585
+}

llmtf_eval_k0_bs1/darumeru_cp_sent_ru_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 128,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "generate"
+    }
+}

llmtf_eval_k0_bs1/darumeru_cp_sent_ru_total.jsonl

@@ -0,0 +1,9 @@
+{
+    "task_name": "darumeru/cp_sent_ru",
+    "results": {
+        "symbol_per_token": 2.8288486051645507,
+        "len": 0.9929878804531707,
+        "lcs": 0.9502007070514518
+    },
+    "leaderboard_result": 0.9929878804531707
+}

llmtf_eval_k0_bs1/darumeru_ruMMLU.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:454e30175faf86134597baf9cad35c4c5a62bdc849eee8d85a27c472c0d729cb
+size 32909111

llmtf_eval_k0_bs1/darumeru_ruMMLU_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 64,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "calculate_tokens_proba"
+    }
+}

llmtf_eval_k0_bs1/darumeru_ruMMLU_total.jsonl

@@ -0,0 +1,7 @@
+{
+    "task_name": "darumeru/ruMMLU",
+    "results": {
+        "acc": 0.5046393295420533
+    },
+    "leaderboard_result": 0.5046393295420533
+}

llmtf_eval_k0_bs1/darumeru_ruOpenBookQA_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 64,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "calculate_tokens_proba"
+    }
+}

llmtf_eval_k0_bs1/darumeru_ruOpenBookQA_total.jsonl

@@ -0,0 +1,8 @@
+{
+    "task_name": "darumeru/ruOpenBookQA",
+    "results": {
+        "acc": 0.6924398625429553,
+        "f1_macro": 0.6928205333186971
+    },
+    "leaderboard_result": 0.6926301979308263
+}

llmtf_eval_k0_bs1/darumeru_ruTiE.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52057ed1300597156320b34738e0b48e92b4112bbd55d3c9b283972df4eb6e15
+size 12832579

llmtf_eval_k0_bs1/darumeru_ruTiE_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 64,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "calculate_tokens_proba"
+    }
+}

llmtf_eval_k0_bs1/darumeru_ruTiE_total.jsonl

@@ -0,0 +1,7 @@
+{
+    "task_name": "darumeru/ruTiE",
+    "results": {
+        "acc": 0.3511627906976744
+    },
+    "leaderboard_result": 0.3511627906976744
+}

llmtf_eval_k0_bs1/darumeru_ruWorldTree_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 64,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "calculate_tokens_proba"
+    }
+}

llmtf_eval_k0_bs1/darumeru_ruWorldTree_total.jsonl

@@ -0,0 +1,8 @@
+{
+    "task_name": "darumeru/ruWorldTree",
+    "results": {
+        "acc": 0.8476190476190476,
+        "f1_macro": 0.8445201637796824
+    },
+    "leaderboard_result": 0.8460696056993651
+}

llmtf_eval_k0_bs1/evaluation_log.txt

@@ -0,0 +1,273 @@
+INFO: 2024-07-13 12:26:10,006: llmtf.base.evaluator: Starting eval on ['darumeru/multiq', 'darumeru/parus', 'darumeru/rcb', 'darumeru/ruopenbookqa', 'darumeru/rutie', 'darumeru/ruworldtree', 'darumeru/rwsd', 'darumeru/use', 'russiannlp/rucola_custom']
+INFO: 2024-07-13 12:26:10,007: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:26:10,007: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:26:10,882: llmtf.base.evaluator: Starting eval on ['nlpcoreteam/rummlu']
+INFO: 2024-07-13 12:26:10,882: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:26:10,882: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:26:10,986: llmtf.base.evaluator: Starting eval on ['darumeru/rummlu']
+INFO: 2024-07-13 12:26:10,986: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:26:10,986: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:26:12,826: llmtf.base.evaluator: Starting eval on ['nlpcoreteam/enmmlu']
+INFO: 2024-07-13 12:26:12,826: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:26:12,826: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:26:13,863: llmtf.base.evaluator: Starting eval on ['daru/treewayabstractive']
+INFO: 2024-07-13 12:26:13,864: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:26:13,864: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:26:13,934: llmtf.base.darumeru/MultiQ: Loading Dataset: 3.93s
+INFO: 2024-07-13 12:26:15,913: llmtf.base.evaluator: Starting eval on ['daru/treewayextractive']
+INFO: 2024-07-13 12:26:15,914: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:26:15,914: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:26:17,900: llmtf.base.evaluator: Starting eval on ['darumeru/cp_sent_ru', 'darumeru/cp_sent_en', 'darumeru/cp_para_ru', 'darumeru/cp_para_en']
+INFO: 2024-07-13 12:26:17,901: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:26:17,901: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:26:18,457: llmtf.base.daru/treewayabstractive: Loading Dataset: 4.59s
+INFO: 2024-07-13 12:26:19,201: llmtf.base.darumeru/ruMMLU: Loading Dataset: 8.21s
+INFO: 2024-07-13 12:26:20,501: llmtf.base.darumeru/cp_sent_ru: Loading Dataset: 2.60s
+INFO: 2024-07-13 12:26:27,467: llmtf.base.daru/treewayextractive: Loading Dataset: 11.55s
+INFO: 2024-07-13 12:28:34,411: llmtf.base.nlpcoreteam/enMMLU: Loading Dataset: 141.58s
+INFO: 2024-07-13 12:28:36,431: llmtf.base.nlpcoreteam/ruMMLU: Loading Dataset: 145.55s
+INFO: 2024-07-13 12:32:28,436: llmtf.base.darumeru/ruMMLU: Processing Dataset: 369.22s
+INFO: 2024-07-13 12:32:28,453: llmtf.base.darumeru/ruMMLU: Results for darumeru/ruMMLU:
+INFO: 2024-07-13 12:32:28,477: llmtf.base.darumeru/ruMMLU: {'acc': 0.5046393295420533}
+INFO: 2024-07-13 12:32:28,511: llmtf.base.evaluator: Ended eval
+INFO: 2024-07-13 12:32:28,516: llmtf.base.evaluator: 
+mean	darumeru/ruMMLU
+0.505	0.505
+INFO: 2024-07-13 12:33:16,551: llmtf.base.daru/treewayextractive: Processing Dataset: 409.08s
+INFO: 2024-07-13 12:33:16,553: llmtf.base.daru/treewayextractive: Results for daru/treewayextractive:
+INFO: 2024-07-13 12:33:16,780: llmtf.base.daru/treewayextractive: {'r-prec': 0.3960751082251082}
+INFO: 2024-07-13 12:33:16,829: llmtf.base.evaluator: Ended eval
+INFO: 2024-07-13 12:33:16,834: llmtf.base.evaluator: 
+mean	daru/treewayextractive	darumeru/ruMMLU
+0.450	0.396	0.505
+INFO: 2024-07-13 12:35:38,012: llmtf.base.darumeru/cp_sent_ru: Processing Dataset: 557.51s
+INFO: 2024-07-13 12:35:38,014: llmtf.base.darumeru/cp_sent_ru: Results for darumeru/cp_sent_ru:
+INFO: 2024-07-13 12:35:38,031: llmtf.base.darumeru/cp_sent_ru: {'symbol_per_token': 2.8288486051645507, 'len': 0.9929878804531707, 'lcs': 0.9502007070514518}
+INFO: 2024-07-13 12:35:38,032: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:35:38,032: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:35:40,708: llmtf.base.darumeru/cp_sent_en: Loading Dataset: 2.67s
+INFO: 2024-07-13 12:36:22,266: llmtf.base.nlpcoreteam/enMMLU: Processing Dataset: 467.84s
+INFO: 2024-07-13 12:36:22,281: llmtf.base.nlpcoreteam/enMMLU: Results for nlpcoreteam/enMMLU:
+INFO: 2024-07-13 12:36:22,320: llmtf.base.nlpcoreteam/enMMLU:                                        metric
+subject                                      
+abstract_algebra                     0.340000
+anatomy                              0.718519
+astronomy                            0.736842
+business_ethics                      0.720000
+clinical_knowledge                   0.735849
+college_biology                      0.791667
+college_chemistry                    0.470000
+college_computer_science             0.590000
+college_mathematics                  0.300000
+college_medicine                     0.653179
+college_physics                      0.480392
+computer_security                    0.760000
+conceptual_physics                   0.565957
+econometrics                         0.517544
+electrical_engineering               0.600000
+elementary_mathematics               0.481481
+formal_logic                         0.523810
+global_facts                         0.410000
+high_school_biology                  0.800000
+high_school_chemistry                0.551724
+high_school_computer_science         0.730000
+high_school_european_history         0.751515
+high_school_geography                0.828283
+high_school_government_and_politics  0.865285
+high_school_macroeconomics           0.633333
+high_school_mathematics              0.366667
+high_school_microeconomics           0.747899
+high_school_physics                  0.423841
+high_school_psychology               0.853211
+high_school_statistics               0.532407
+high_school_us_history               0.828431
+high_school_world_history            0.835443
+human_aging                          0.721973
+human_sexuality                      0.778626
+international_law                    0.760331
+jurisprudence                        0.796296
+logical_fallacies                    0.779141
+machine_learning                     0.455357
+management                           0.805825
+marketing                            0.893162
+medical_genetics                     0.780000
+miscellaneous                        0.837803
+moral_disputes                       0.690751
+moral_scenarios                      0.289385
+nutrition                            0.764706
+philosophy                           0.720257
+prehistory                           0.709877
+professional_accounting              0.531915
+professional_law                     0.479140
+professional_medicine                0.731618
+professional_psychology              0.674837
+public_relations                     0.654545
+security_studies                     0.714286
+sociology                            0.825871
+us_foreign_policy                    0.890000
+virology                             0.487952
+world_religions                      0.824561
+INFO: 2024-07-13 12:36:22,327: llmtf.base.nlpcoreteam/enMMLU:                                    metric
+subject                                  
+STEM                             0.554241
+humanities                       0.691457
+other (business, health, misc.)  0.699464
+social sciences                  0.748643
+INFO: 2024-07-13 12:36:22,379: llmtf.base.nlpcoreteam/enMMLU: {'acc': 0.6734513449759852}
+INFO: 2024-07-13 12:36:22,409: llmtf.base.evaluator: Ended eval
+INFO: 2024-07-13 12:36:22,416: llmtf.base.evaluator: 
+mean	daru/treewayextractive	darumeru/cp_sent_ru	darumeru/ruMMLU	nlpcoreteam/enMMLU
+0.642	0.396	0.993	0.505	0.673
+INFO: 2024-07-13 12:37:10,439: llmtf.base.nlpcoreteam/ruMMLU: Processing Dataset: 514.01s
+INFO: 2024-07-13 12:37:10,457: llmtf.base.nlpcoreteam/ruMMLU: Results for nlpcoreteam/ruMMLU:
+INFO: 2024-07-13 12:37:10,498: llmtf.base.nlpcoreteam/ruMMLU:                                        metric
+subject                                      
+abstract_algebra                     0.290000
+anatomy                              0.459259
+astronomy                            0.657895
+business_ethics                      0.600000
+clinical_knowledge                   0.562264
+college_biology                      0.548611
+college_chemistry                    0.400000
+college_computer_science             0.470000
+college_mathematics                  0.330000
+college_medicine                     0.497110
+college_physics                      0.333333
+computer_security                    0.570000
+conceptual_physics                   0.493617
+econometrics                         0.342105
+electrical_engineering               0.531034
+elementary_mathematics               0.412698
+formal_logic                         0.380952
+global_facts                         0.350000
+high_school_biology                  0.635484
+high_school_chemistry                0.428571
+high_school_computer_science         0.620000
+high_school_european_history         0.715152
+high_school_geography                0.656566
+high_school_government_and_politics  0.595855
+high_school_macroeconomics           0.517949
+high_school_mathematics              0.348148
+high_school_microeconomics           0.495798
+high_school_physics                  0.350993
+high_school_psychology               0.667890
+high_school_statistics               0.458333
+high_school_us_history               0.661765
+high_school_world_history            0.708861
+human_aging                          0.556054
+human_sexuality                      0.664122
+international_law                    0.702479
+jurisprudence                        0.592593
+logical_fallacies                    0.527607
+machine_learning                     0.339286
+management                           0.669903
+marketing                            0.700855
+medical_genetics                     0.570000
+miscellaneous                        0.646232
+moral_disputes                       0.554913
+moral_scenarios                      0.248045
+nutrition                            0.594771
+philosophy                           0.565916
+prehistory                           0.558642
+professional_accounting              0.386525
+professional_law                     0.362451
+professional_medicine                0.522059
+professional_psychology              0.480392
+public_relations                     0.563636
+security_studies                     0.620408
+sociology                            0.696517
+us_foreign_policy                    0.770000
+virology                             0.415663
+world_religions                      0.690058
+INFO: 2024-07-13 12:37:10,507: llmtf.base.nlpcoreteam/ruMMLU:                                    metric
+subject                                  
+STEM                             0.456556
+humanities                       0.559187
+other (business, health, misc.)  0.537907
+social sciences                  0.589270
+INFO: 2024-07-13 12:37:10,515: llmtf.base.nlpcoreteam/ruMMLU: {'acc': 0.5357299468552112}
+INFO: 2024-07-13 12:37:10,547: llmtf.base.evaluator: Ended eval
+INFO: 2024-07-13 12:37:10,554: llmtf.base.evaluator: 
+mean	daru/treewayextractive	darumeru/cp_sent_ru	darumeru/ruMMLU	nlpcoreteam/enMMLU	nlpcoreteam/ruMMLU
+0.621	0.396	0.993	0.505	0.673	0.536
+INFO: 2024-07-13 12:37:40,183: llmtf.base.darumeru/MultiQ: Processing Dataset: 686.25s
+INFO: 2024-07-13 12:37:40,186: llmtf.base.darumeru/MultiQ: Results for darumeru/MultiQ:
+INFO: 2024-07-13 12:37:40,191: llmtf.base.darumeru/MultiQ: {'f1': 0.33947318357368605, 'em': 0.21414913957934992}
+INFO: 2024-07-13 12:37:40,196: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:37:40,196: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:37:42,624: llmtf.base.darumeru/PARus: Loading Dataset: 2.43s
+INFO: 2024-07-13 12:37:48,888: llmtf.base.darumeru/PARus: Processing Dataset: 6.26s
+INFO: 2024-07-13 12:37:48,890: llmtf.base.darumeru/PARus: Results for darumeru/PARus:
+INFO: 2024-07-13 12:37:48,902: llmtf.base.darumeru/PARus: {'acc': 0.66}
+INFO: 2024-07-13 12:37:48,904: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:37:48,904: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:37:51,352: llmtf.base.darumeru/RCB: Loading Dataset: 2.45s
+INFO: 2024-07-13 12:37:59,057: llmtf.base.darumeru/RCB: Processing Dataset: 7.69s
+INFO: 2024-07-13 12:37:59,059: llmtf.base.darumeru/RCB: Results for darumeru/RCB:
+INFO: 2024-07-13 12:37:59,078: llmtf.base.darumeru/RCB: {'acc': 0.5, 'f1_macro': 0.43018975381906}
+INFO: 2024-07-13 12:37:59,080: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:37:59,080: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:38:01,863: llmtf.base.darumeru/ruOpenBookQA: Loading Dataset: 2.78s
+INFO: 2024-07-13 12:39:16,087: llmtf.base.darumeru/ruOpenBookQA: Processing Dataset: 74.22s
+INFO: 2024-07-13 12:39:16,089: llmtf.base.darumeru/ruOpenBookQA: Results for darumeru/ruOpenBookQA:
+INFO: 2024-07-13 12:39:16,102: llmtf.base.darumeru/ruOpenBookQA: {'acc': 0.6924398625429553, 'f1_macro': 0.6928205333186971}
+INFO: 2024-07-13 12:39:16,109: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:39:16,109: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:39:23,241: llmtf.base.darumeru/ruTiE: Loading Dataset: 7.13s
+INFO: 2024-07-13 12:43:16,950: llmtf.base.darumeru/cp_sent_en: Processing Dataset: 456.24s
+INFO: 2024-07-13 12:43:16,970: llmtf.base.darumeru/cp_sent_en: Results for darumeru/cp_sent_en:
+INFO: 2024-07-13 12:43:16,991: llmtf.base.darumeru/cp_sent_en: {'symbol_per_token': 4.424738337449315, 'len': 0.9996416196590585, 'lcs': 0.9958136839407484}
+INFO: 2024-07-13 12:43:16,992: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:43:16,992: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:43:19,141: llmtf.base.darumeru/cp_para_ru: Loading Dataset: 2.15s
+INFO: 2024-07-13 12:43:54,391: llmtf.base.darumeru/ruTiE: Processing Dataset: 271.14s
+INFO: 2024-07-13 12:43:54,392: llmtf.base.darumeru/ruTiE: Results for darumeru/ruTiE:
+INFO: 2024-07-13 12:43:54,420: llmtf.base.darumeru/ruTiE: {'acc': 0.3511627906976744}
+INFO: 2024-07-13 12:43:54,423: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:43:54,423: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:43:57,142: llmtf.base.darumeru/ruWorldTree: Loading Dataset: 2.72s
+INFO: 2024-07-13 12:44:00,566: llmtf.base.darumeru/ruWorldTree: Processing Dataset: 3.42s
+INFO: 2024-07-13 12:44:00,583: llmtf.base.darumeru/ruWorldTree: Results for darumeru/ruWorldTree:
+INFO: 2024-07-13 12:44:00,589: llmtf.base.darumeru/ruWorldTree: {'acc': 0.8476190476190476, 'f1_macro': 0.8445201637796824}
+INFO: 2024-07-13 12:44:00,590: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:44:00,590: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:44:02,920: llmtf.base.darumeru/RWSD: Loading Dataset: 2.33s
+INFO: 2024-07-13 12:44:10,009: llmtf.base.darumeru/RWSD: Processing Dataset: 7.09s
+INFO: 2024-07-13 12:44:10,025: llmtf.base.darumeru/RWSD: Results for darumeru/RWSD:
+INFO: 2024-07-13 12:44:10,029: llmtf.base.darumeru/RWSD: {'acc': 0.5441176470588235}
+INFO: 2024-07-13 12:44:10,030: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:44:10,030: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:44:12,714: llmtf.base.darumeru/USE: Loading Dataset: 2.68s
+INFO: 2024-07-13 12:55:27,344: llmtf.base.darumeru/USE: Processing Dataset: 674.63s
+INFO: 2024-07-13 12:55:27,350: llmtf.base.darumeru/USE: Results for darumeru/USE:
+INFO: 2024-07-13 12:55:27,461: llmtf.base.darumeru/USE: {'grade_norm': 0.07941176470588233}
+INFO: 2024-07-13 12:55:27,464: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:55:27,464: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:55:30,023: llmtf.base.darumeru/cp_para_ru: Processing Dataset: 730.88s
+INFO: 2024-07-13 12:55:30,025: llmtf.base.darumeru/cp_para_ru: Results for darumeru/cp_para_ru:
+INFO: 2024-07-13 12:55:30,028: llmtf.base.darumeru/cp_para_ru: {'symbol_per_token': 2.965959526577639, 'len': 0.9943389906706583, 'lcs': 0.9089360152526831}
+INFO: 2024-07-13 12:55:30,029: llmtf.base.hfmodel: Updated generation_config.eos_token_id: [128001, 128009]
+INFO: 2024-07-13 12:55:30,029: llmtf.base.hfmodel: Updated generation_config.stop_strings: []
+INFO: 2024-07-13 12:55:32,480: llmtf.base.russiannlp/rucola_custom: Loading Dataset: 5.02s
+INFO: 2024-07-13 12:55:59,692: llmtf.base.darumeru/cp_para_en: Loading Dataset: 29.66s
+INFO: 2024-07-13 12:56:23,955: llmtf.base.daru/treewayabstractive: Processing Dataset: 1805.50s
+INFO: 2024-07-13 12:56:23,957: llmtf.base.daru/treewayabstractive: Results for daru/treewayabstractive:
+INFO: 2024-07-13 12:56:23,966: llmtf.base.daru/treewayabstractive: {'rouge1': 0.35429775115253526, 'rouge2': 0.12889173695955833}
+INFO: 2024-07-13 12:56:23,968: llmtf.base.evaluator: Ended eval
+INFO: 2024-07-13 12:56:24,039: llmtf.base.evaluator: 
+mean	daru/treewayabstractive	daru/treewayextractive	darumeru/MultiQ	darumeru/PARus	darumeru/RCB	darumeru/RWSD	darumeru/USE	darumeru/cp_para_ru	darumeru/cp_sent_en	darumeru/cp_sent_ru	darumeru/ruMMLU	darumeru/ruOpenBookQA	darumeru/ruTiE	darumeru/ruWorldTree	nlpcoreteam/enMMLU	nlpcoreteam/ruMMLU
+0.573	0.242	0.396	0.277	0.660	0.465	0.544	0.079	0.909	1.000	0.993	0.505	0.693	0.351	0.846	0.673	0.536
+INFO: 2024-07-13 12:56:59,954: llmtf.base.russiannlp/rucola_custom: Processing Dataset: 87.47s
+INFO: 2024-07-13 12:56:59,973: llmtf.base.russiannlp/rucola_custom: Results for russiannlp/rucola_custom:
+INFO: 2024-07-13 12:57:00,001: llmtf.base.russiannlp/rucola_custom: {'acc': 0.7068532472192322, 'mcc': 0.2623100586905413}
+INFO: 2024-07-13 12:57:00,005: llmtf.base.evaluator: Ended eval
+INFO: 2024-07-13 12:57:00,016: llmtf.base.evaluator: 
+mean	daru/treewayabstractive	daru/treewayextractive	darumeru/MultiQ	darumeru/PARus	darumeru/RCB	darumeru/RWSD	darumeru/USE	darumeru/cp_para_ru	darumeru/cp_sent_en	darumeru/cp_sent_ru	darumeru/ruMMLU	darumeru/ruOpenBookQA	darumeru/ruTiE	darumeru/ruWorldTree	nlpcoreteam/enMMLU	nlpcoreteam/ruMMLU	russiannlp/rucola_custom
+0.568	0.242	0.396	0.277	0.660	0.465	0.544	0.079	0.909	1.000	0.993	0.505	0.693	0.351	0.846	0.673	0.536	0.485
+INFO: 2024-07-13 13:05:43,824: llmtf.base.darumeru/cp_para_en: Processing Dataset: 584.13s
+INFO: 2024-07-13 13:05:43,847: llmtf.base.darumeru/cp_para_en: Results for darumeru/cp_para_en:
+INFO: 2024-07-13 13:05:43,850: llmtf.base.darumeru/cp_para_en: {'symbol_per_token': 4.463061170262149, 'len': 0.9941296296409974, 'lcs': 0.9527227031116661}
+INFO: 2024-07-13 13:05:43,851: llmtf.base.evaluator: Ended eval
+INFO: 2024-07-13 13:05:43,862: llmtf.base.evaluator: 
+mean	daru/treewayabstractive	daru/treewayextractive	darumeru/MultiQ	darumeru/PARus	darumeru/RCB	darumeru/RWSD	darumeru/USE	darumeru/cp_para_en	darumeru/cp_para_ru	darumeru/cp_sent_en	darumeru/cp_sent_ru	darumeru/ruMMLU	darumeru/ruOpenBookQA	darumeru/ruTiE	darumeru/ruWorldTree	nlpcoreteam/enMMLU	nlpcoreteam/ruMMLU	russiannlp/rucola_custom
+0.589	0.242	0.396	0.277	0.660	0.465	0.544	0.079	0.953	0.909	1.000	0.993	0.505	0.693	0.351	0.846	0.673	0.536	0.485

llmtf_eval_k0_bs1/evaluation_results.txt

@@ -0,0 +1,2 @@
+mean	daru/treewayabstractive	daru/treewayextractive	darumeru/MultiQ	darumeru/PARus	darumeru/RCB	darumeru/RWSD	darumeru/USE	darumeru/cp_para_en	darumeru/cp_para_ru	darumeru/cp_sent_en	darumeru/cp_sent_ru	darumeru/ruMMLU	darumeru/ruOpenBookQA	darumeru/ruTiE	darumeru/ruWorldTree	nlpcoreteam/enMMLU	nlpcoreteam/ruMMLU	russiannlp/rucola_custom
+0.589	0.242	0.396	0.277	0.660	0.465	0.544	0.079	0.953	0.909	1.000	0.993	0.505	0.693	0.351	0.846	0.673	0.536	0.485

llmtf_eval_k0_bs1/nlpcoreteam_enMMLU.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a3488fe16dd4c6cc5496a219929273e2746e659cceba1baa45750cf002c05f5
+size 38085342

llmtf_eval_k0_bs1/nlpcoreteam_enMMLU_params.jsonl

@@ -0,0 +1,57 @@
+{
+    "custom_generation_config": null,
+    "model_params": {
+        "model_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+        "generation_config": {
+            "bos_token_id": 128000,
+            "do_sample": true,
+            "eos_token_id": [
+                128001,
+                128009
+            ],
+            "max_length": 8192,
+            "max_new_tokens": 64,
+            "pad_token_id": 128001,
+            "stop_strings": [],
+            "temperature": 0.1,
+            "top_k": 40,
+            "top_p": 0.9,
+            "transformers_version": "4.38.2",
+            "trust_remote_code": [
+                false
+            ]
+        },
+        "conversation_template": {
+            "system_prompt": "",
+            "system_message_template": "",
+            "user_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|>",
+            "bot_message_template_incomplete": "<|start_header_id|>{role}<|end_header_id|>\n\n{content}",
+            "user_role": "user",
+            "bot_role": "assistant",
+            "system_role": "system",
+            "global_prefix": "<|begin_of_text|>",
+            "suffix": "<|start_header_id|>assistant<|end_header_id|>\n\n",
+            "add_special_tokens": false,
+            "eos_token": "<|eot_id|>"
+        },
+        "load_in_8bit": false,
+        "torch_dtype": "auto",
+        "use_flash_attention_2": true,
+        "device_map": "cuda:0",
+        "use_fast_tokenizer": true,
+        "leading_space": false,
+        "space_token": null,
+        "trust_remote_code": [
+            false
+        ],
+        "max_model_len": 8192
+    },
+    "task_params": {
+        "max_len": 4000,
+        "few_shot_count": 0,
+        "batch_size": 1,
+        "max_sample_per_dataset": 10000000000000,
+        "method": "calculate_tokens_proba"
+    }
+}