Upload ray_tune_logs/result.json with huggingface_hub

ray_tune_logs/result.json

@@ -1 +1 @@
-{"samsum/rouge1": 0.055488210861802195, "samsum/rouge2": 0.0148884485738399, "samsum/rougeL": 0.041491939300558646, "samsum/rougeLsum": 0.042507458030167145, "samsum/bertscore_precision": 0.6166475216547648, "samsum/bertscore_recall": 0.6834982385238012, "samsum/bertscore_f1": 0.6476827760537466, "cnndm/rouge1": 0.13016523589045112, "cnndm/rouge2": 0.041991571069563326, "cnndm/rougeL": 0.10993746712966011, "cnndm/rougeLsum": 0.10715990114906183, "cnndm/bertscore_precision": 0.681688129901886, "cnndm/bertscore_recall": 0.7389524877071381, "cnndm/bertscore_f1": 0.7080690910418829, "gigaword/rouge1": 0.016626713906933873, "gigaword/rouge2": 0.0034357494717621785, "gigaword/rougeL": 0.01630322037591579, "gigaword/rougeLsum": 0.015539180130949517, "gigaword/bertscore_precision": 0.5684015327692031, "gigaword/bertscore_recall": 0.6505913469195366, "gigaword/bertscore_f1": 0.6057714141905308, "xsum/rouge1": 0.10035712986483537, "xsum/rouge2": 0.018761301067129577, "xsum/rougeL": 0.08373656314284877, "xsum/rougeLsum": 0.08393338429555226, "xsum/bertscore_precision": 0.698464055856069, "xsum/bertscore_recall": 0.7191349665323893, "xsum/bertscore_f1": 0.7044563740491867, "eval_agg/avg_all_rougef": 0.05514521714131448, "eval_agg/avg_all_bertf": 0.6664949138338367, "eval_agg/avg_all": 0.3608200654875756, "num_rl_rollout": 0, "lm_epoch": 0, "rl_epoch": 0, "step": 0, "total_data_token": 0, "total_rl_token": 0, "total_lm_token": 0, "total_token": 0, "completed_steps": 0, "tune_objective": 0.8010626369032927, "timestamp": 1771899561, "checkpoint_dir_name": null, "done": true, "training_iteration": 1, "trial_id": "b01e2ab8", "date": "2026-02-24_10-19-21", "time_this_iter_s": 57.88321614265442, "time_total_s": 57.88321614265442, "pid": 3657345, "hostname": "lagoon", "node_ip": "10.2.1.31", "config": {"train_loop_config": {"dataset_name": "nbtpj/summ_ds_train", "dataset_config_name": null, "train_split_name": "merge36_cnndmsamsumxsum", "text_template": "{text}\nTL;DR: {summary}", "label_col": "summary", "freeze_role2": false, "only_train_role1": false, "model_name_or_path": "gpt2", "ref_role1_name_or_path": "gpt2", "ref_role2_name_or_path": "gpt2", "pretrained_role2_name_or_path": "none", "config_name": null, "vectorizer_path": "/common/home/users/m/mq.nguyen.2023/testcode/SAC_LM/module9_clmv2/vectorizer/wikitext103_tfidf_full.joblib", "tokenizer_name": null, "use_slow_tokenizer": false, "per_device_train_batch_size": 4, "per_device_query_rollout_batch_size": 32, "per_device_eval_batch_size": 4, "vllm_vram_ratio": 0.3, "learning_rate": 3e-07, "grad_norm": 0.5, "weight_decay": 1e-05, "max_train_steps": 40000, "max_train_rollouts": 100000, "gradient_accumulation_steps": 1, "lr_scheduler_type": "constant", "num_warmup_steps": 200, "seed": 0, "model_type": null, "block_size": 1024, "mini_epoch": 1, "rollout_game": "baseline3v2", "rl_algo": "off_policy", "constraint_type": "kl", "clamp_update": false, "rl_w": 1.0, "lm_w": 1.0, "n_generate": 4, "n_augment": 0, "gradient_checkpoint": false, "group_relative_norm": false, "sample_config": {"do_sample": true, "min_new_tokens": 20, "temperature": 1.0}, "inference_config": {"do_sample": true, "temperature": 0.0, "min_new_tokens": 5, "max_new_tokens": 250}, "rollout_config": {"accuracy_w": 0.03744666535741567, "len_pen": 1.0, "accuracy_w2": 60.07249475906205, "len_pen2": 1.0, "threshold": 0.011058510725696462, "similarity_fn": "rouge", "n_gen1": 1, "n_gen2": 1}, "ent_coef": 0.0001, "beta_coef": "0.0", "prompt_0": "{text}", "prompt_1": "{text}\nTL;DR: ", "prompt_2": "Given the text: {role1_output}\nReconstruct the summarized text to the detailed:", "prompt_eval": "{text}\nTL;DR:", "epsilon": 0.2, "a2g_norm": true, "vllm_sleep": true, "lora": true, "need_attn_mask": true, "gamma": 0.95, "trust_remote_code": true, "test_glue": false, "test_clm": false, "causal_model": true, "test_gen": true, "log_rollout_txt": true, "trunc_eval": 256, "buffer_max_size": 20000, "trunc_evals": ["cnndm___12", "samsum___12", "xsum___12", "gigaword___200", "duc___50"], "use_deepspeed": false, "zero_config": 2, "log_interval": "5m", "eval_interval": "100", "checkpoint_interval": "100", "lm_fraction": -1.0, "push_to_hub": null, "keep_eval_size": false, "mixed_precision": "bf16", "tune_metrics": ["cnndm/rouge1___1.0", "cnndm/bertscore_f1___0.25", "samsum/rouge1___1.0", "samsum/bertscore_f1___0.25", "xsum/rouge1___1.0", "xsum/bertscore_f1___0.25"], "base_path": "/common/home/users/m/mq.nguyen.2023/testcode/SAC_LM/module9_clmv2", "script": "/common/home/users/m/mq.nguyen.2023/testcode/SAC_LM/module9_clmv2/execute/utune/debug.py", "train_from_raw": true, "world_size": 1, "cpu_per_worker": 7, "gpu_per_worker": 1}}, "time_since_restore": 57.88321614265442, "iterations_since_restore": 1}
+{"cnndm/rouge1": 0.13016523589045112, "cnndm/rouge2": 0.041991571069563326, "cnndm/rougeL": 0.10993746712966011, "cnndm/rougeLsum": 0.10715990114906183, "cnndm/bertscore_precision": 0.681688129901886, "cnndm/bertscore_recall": 0.7389524976412455, "cnndm/bertscore_f1": 0.7080691009759903, "samsum/rouge1": 0.055488210861802195, "samsum/rouge2": 0.0148884485738399, "samsum/rougeL": 0.041491939300558646, "samsum/rougeLsum": 0.042507458030167145, "samsum/bertscore_precision": 0.6166475266218185, "samsum/bertscore_recall": 0.6834982484579086, "samsum/bertscore_f1": 0.6476827810208002, "gigaword/rouge1": 0.016626713906933873, "gigaword/rouge2": 0.0034357494717621785, "gigaword/rougeL": 0.01630322037591579, "gigaword/rougeLsum": 0.015539180130949517, "gigaword/bertscore_precision": 0.5684015312790871, "gigaword/bertscore_recall": 0.6505913436412811, "gigaword/bertscore_f1": 0.6057714119553566, "xsum/rouge1": 0.10035712986483537, "xsum/rouge2": 0.018761301067129577, "xsum/rougeL": 0.08373656314284877, "xsum/rougeLsum": 0.08393338429555226, "xsum/bertscore_precision": 0.698464055856069, "xsum/bertscore_recall": 0.719134971499443, "xsum/bertscore_f1": 0.7044563889503479, "eval_agg/avg_all_rougef": 0.05514521714131448, "eval_agg/avg_all_bertf": 0.6664949207256237, "eval_agg/avg_all": 0.3608200689334691, "num_rl_rollout": 0, "lm_epoch": 0, "rl_epoch": 0, "step": 0, "total_data_token": 0, "total_rl_token": 0, "total_lm_token": 0, "total_token": 0, "completed_steps": 0, "tune_objective": 0.8010626443538733, "timestamp": 1771899563, "checkpoint_dir_name": null, "done": true, "training_iteration": 1, "trial_id": "6ad1f914", "date": "2026-02-24_10-19-23", "time_this_iter_s": 59.47291588783264, "time_total_s": 59.47291588783264, "pid": 3657602, "hostname": "lagoon", "node_ip": "10.2.1.31", "config": {"train_loop_config": {"dataset_name": "nbtpj/summ_ds_train", "dataset_config_name": null, "train_split_name": "merge36_cnndmsamsumxsum", "text_template": "{text}\nTL;DR: {summary}", "label_col": "summary", "freeze_role2": false, "only_train_role1": false, "model_name_or_path": "gpt2", "ref_role1_name_or_path": "gpt2", "ref_role2_name_or_path": "gpt2", "pretrained_role2_name_or_path": "none", "config_name": null, "vectorizer_path": "/common/home/users/m/mq.nguyen.2023/testcode/SAC_LM/module9_clmv2/vectorizer/wikitext103_tfidf_full.joblib", "tokenizer_name": null, "use_slow_tokenizer": false, "per_device_train_batch_size": 4, "per_device_query_rollout_batch_size": 32, "per_device_eval_batch_size": 4, "vllm_vram_ratio": 0.3, "learning_rate": 3e-07, "grad_norm": 0.5, "weight_decay": 1e-05, "max_train_steps": 40000, "max_train_rollouts": 100000, "gradient_accumulation_steps": 1, "lr_scheduler_type": "constant", "num_warmup_steps": 200, "seed": 0, "model_type": null, "block_size": 1024, "mini_epoch": 1, "rollout_game": "baseline3v2", "rl_algo": "off_policy", "constraint_type": "kl", "clamp_update": false, "rl_w": 1.0, "lm_w": 1.0, "n_generate": 4, "n_augment": 0, "gradient_checkpoint": false, "group_relative_norm": true, "sample_config": {"do_sample": true, "min_new_tokens": 20, "temperature": 1.0}, "inference_config": {"do_sample": true, "temperature": 0.0, "min_new_tokens": 5, "max_new_tokens": 250}, "rollout_config": {"accuracy_w": 0.49501595537331966, "len_pen": 1.0, "accuracy_w2": 3.833332156156665, "len_pen2": 1.0, "threshold": 0.007501954443620123, "similarity_fn": "rouge", "n_gen1": 2, "n_gen2": 2}, "ent_coef": 0.0001, "beta_coef": "0.0", "prompt_0": "{text}", "prompt_1": "{text}\nTL;DR: ", "prompt_2": "Given the text: {role1_output}\nReconstruct the summarized text to the detailed:", "prompt_eval": "{text}\nTL;DR:", "epsilon": 0.2, "a2g_norm": true, "vllm_sleep": true, "lora": true, "need_attn_mask": true, "gamma": 0.95, "trust_remote_code": true, "test_glue": false, "test_clm": false, "causal_model": true, "test_gen": true, "log_rollout_txt": true, "trunc_eval": 256, "buffer_max_size": 20000, "trunc_evals": ["cnndm___12", "samsum___12", "xsum___12", "gigaword___200", "duc___50"], "use_deepspeed": false, "zero_config": 2, "log_interval": "5m", "eval_interval": "100", "checkpoint_interval": "100", "lm_fraction": -1.0, "push_to_hub": null, "keep_eval_size": false, "mixed_precision": "bf16", "tune_metrics": ["cnndm/rouge1___1.0", "cnndm/bertscore_f1___0.25", "samsum/rouge1___1.0", "samsum/bertscore_f1___0.25", "xsum/rouge1___1.0", "xsum/bertscore_f1___0.25"], "base_path": "/common/home/users/m/mq.nguyen.2023/testcode/SAC_LM/module9_clmv2", "script": "/common/home/users/m/mq.nguyen.2023/testcode/SAC_LM/module9_clmv2/execute/utune/debug.py", "train_from_raw": true, "world_size": 1, "cpu_per_worker": 7, "gpu_per_worker": 1}}, "time_since_restore": 59.47291588783264, "iterations_since_restore": 1}