End of training

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: Gensyn/Qwen2.5-0.5B-Instruct
+base_model: Gensyn/Qwen2.5-1.5B-Instruct
 library_name: transformers
 model_name: gensyn-checkpoints-flexible_scruffy_condor
 tags:
@@ -8,13 +8,14 @@ tags:
 - grpo
 - gensyn
 - I am flexible scruffy condor
+- unsloth
 - trl
 licence: license
 ---
 
 # Model Card for gensyn-checkpoints-flexible_scruffy_condor
 
-This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
+This model is a fine-tuned version of [Gensyn/Qwen2.5-1.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-1.5B-Instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -39,7 +40,7 @@ This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing
 
 - TRL: 0.15.2
 - Transformers: 4.51.3
-- Pytorch: 2.5.1
+- Pytorch: 2.6.0
 - Datasets: 3.5.0
 - Tokenizers: 0.21.1
 

adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Gensyn/Qwen2.5-1.5B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "down_proj",
+    "up_proj",
+    "gate_proj",
+    "o_proj",
+    "q_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c651b67e0f28f123c7d96183c509f4f4095ca7a070633dd73ed4b26d806a132
+size 73911112

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 2.8217583894729616e-05,
-    "train_runtime": 239.7783,
-    "train_samples": 28,
-    "train_samples_per_second": 1.335,
-    "train_steps_per_second": 0.083
+    "train_loss": 9.213837813604187e-07,
+    "train_runtime": 1717.4399,
+    "train_samples": 100,
+    "train_samples_per_second": 0.186,
+    "train_steps_per_second": 0.012
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 2.8217583894729616e-05,
-    "train_runtime": 239.7783,
-    "train_samples": 28,
-    "train_samples_per_second": 1.335,
-    "train_steps_per_second": 0.083
+    "train_loss": 9.213837813604187e-07,
+    "train_runtime": 1717.4399,
+    "train_samples": 100,
+    "train_samples_per_second": 0.186,
+    "train_steps_per_second": 0.012
 }

trainer_state.json

@@ -2,7 +2,7 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 5.0,
+  "epoch": 0.8,
   "eval_steps": 500,
   "global_step": 20,
   "is_hyper_param_search": false,
@@ -10,209 +10,209 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "completion_length": 351.1875,
-      "epoch": 0.5714285714285714,
-      "grad_norm": 4.696744918823242,
-      "kl": 0.0,
+      "completion_length": 524.6875,
+      "epoch": 0.08,
+      "grad_norm": 0.3149994909763336,
+      "kl": 0.0007895925155025907,
       "learning_rate": 5e-07,
-      "loss": -0.0,
-      "reward": 1.8700803834944963,
-      "reward_std": 1.1751239793375134,
-      "rewards/concensus_correctness_reward_func": 0.3256249986588955,
-      "rewards/consensus_reward_func": 0.5,
+      "loss": 0.0,
+      "reward": 0.84690111130476,
+      "reward_std": 0.6340070366859436,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.4988928958773613,
+      "rewards/question_recreation_reward_func": 0.3991198819130659,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.5455625092145056,
+      "rewards/xmlcount_reward_func": 0.44778124429285526,
       "step": 2
     },
     {
-      "completion_length": 276.5416666666667,
-      "epoch": 1.0,
-      "grad_norm": 4.756100654602051,
-      "kl": 0.001327126282073247,
+      "completion_length": 287.40625,
+      "epoch": 0.16,
+      "grad_norm": 0.8252567648887634,
+      "kl": 0.0008640004525659606,
       "learning_rate": 4.864543104251586e-07,
       "loss": 0.0,
-      "reward": 2.3633016645908356,
-      "reward_std": 0.722859085459883,
-      "rewards/concensus_correctness_reward_func": 0.4736666629711787,
-      "rewards/consensus_reward_func": 0.5,
+      "reward": 1.479073416441679,
+      "reward_std": 1.898745086044073,
+      "rewards/concensus_correctness_reward_func": 0.625,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.49659330933354795,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.4101983695290983,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.8930416852235794,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.30324999522417784,
       "step": 4
     },
     {
-      "completion_length": 293.125,
-      "epoch": 1.5714285714285714,
-      "grad_norm": 6.781271934509277,
-      "kl": 0.004252150494721718,
+      "completion_length": 512.71875,
+      "epoch": 0.24,
+      "grad_norm": 0.46688419580459595,
+      "kl": 0.000702431112586055,
       "learning_rate": 4.472851273490984e-07,
       "loss": 0.0,
-      "reward": 2.769148699939251,
-      "reward_std": 0.910884540528059,
-      "rewards/concensus_correctness_reward_func": 0.6645000018179417,
-      "rewards/consensus_reward_func": 0.8125,
+      "reward": 1.4388332553207874,
+      "reward_std": 1.226584866642952,
+      "rewards/concensus_correctness_reward_func": 0.15937500074505806,
+      "rewards/consensus_reward_func": 0.1875,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.5405236706137657,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.5340207722038031,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.7516250051558018,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.3548125009983778,
       "step": 6
     },
     {
-      "completion_length": 423.9583333333333,
-      "epoch": 2.0,
-      "grad_norm": 3.4789960384368896,
-      "kl": 0.0026184429346661395,
+      "completion_length": 418.65625,
+      "epoch": 0.32,
+      "grad_norm": 0.518238365650177,
+      "kl": 0.0008219250739784911,
       "learning_rate": 3.867370395306068e-07,
       "loss": 0.0,
-      "reward": 2.372594674428304,
-      "reward_std": 1.0563052495320637,
-      "rewards/concensus_correctness_reward_func": 0.3671666607260704,
-      "rewards/consensus_reward_func": 0.6666666666666666,
+      "reward": 0.9953836053609848,
+      "reward_std": 0.8189685866236687,
+      "rewards/concensus_correctness_reward_func": 0.015375000424683094,
+      "rewards/consensus_reward_func": 0.125,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.6337196112920841,
+      "rewards/question_recreation_reward_func": 0.5274773649871349,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.7050416593750318,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.31190625159069896,
       "step": 8
     },
     {
-      "completion_length": 279.96875,
-      "epoch": 2.571428571428571,
-      "grad_norm": 7.827857494354248,
-      "kl": 0.016727088644984178,
+      "completion_length": 444.59375,
+      "epoch": 0.4,
+      "grad_norm": 8.409585952758789,
+      "kl": 0.0009162062997347675,
       "learning_rate": 3.1137137178519977e-07,
       "loss": 0.0,
-      "reward": 2.6220436356961727,
-      "reward_std": 1.3839910216629505,
-      "rewards/concensus_correctness_reward_func": 0.4554375000298023,
-      "rewards/consensus_reward_func": 0.75,
+      "reward": 1.1158432438969612,
+      "reward_std": 0.6383559834212065,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.4946998809464276,
+      "rewards/question_recreation_reward_func": 0.608686987310648,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.906281266361475,
+      "rewards/strict_format_reward_func": 0.03125,
+      "rewards/xmlcount_reward_func": 0.47590626776218414,
       "step": 10
     },
     {
-      "completion_length": 269.5,
-      "epoch": 3.0,
-      "grad_norm": 3.278268337249756,
-      "kl": 0.009888109013748666,
+      "completion_length": 495.53125,
+      "epoch": 0.48,
+      "grad_norm": 0.519533634185791,
+      "kl": 0.0007467774557881057,
       "learning_rate": 2.2935516363191693e-07,
       "loss": 0.0,
-      "reward": 2.2351842870314917,
-      "reward_std": 0.7502500675618649,
-      "rewards/concensus_correctness_reward_func": 0.32758333285649616,
-      "rewards/consensus_reward_func": 0.4166666666666667,
+      "reward": 0.7604115232825279,
+      "reward_std": 1.1575603373348713,
+      "rewards/concensus_correctness_reward_func": 0.06250000093132257,
+      "rewards/consensus_reward_func": 0.125,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.5956842492644986,
+      "rewards/question_recreation_reward_func": 0.31769278924912214,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.020833333333333332,
-      "rewards/xmlcount_reward_func": 0.8744166692097982,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.2552187405526638,
       "step": 12
     },
     {
-      "completion_length": 307.875,
-      "epoch": 3.571428571428571,
-      "grad_norm": 5.234099388122559,
-      "kl": 0.009493237361311913,
+      "completion_length": 405.96875,
+      "epoch": 0.56,
+      "grad_norm": 0.505827009677887,
+      "kl": 0.0008590476791141555,
       "learning_rate": 1.4957614383675767e-07,
       "loss": 0.0,
-      "reward": 2.608593128621578,
-      "reward_std": 1.3945186883211136,
-      "rewards/concensus_correctness_reward_func": 0.43618750013411045,
-      "rewards/consensus_reward_func": 0.5625,
+      "reward": 0.7179178791120648,
+      "reward_std": 0.6347920149564743,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.6657494021928869,
+      "rewards/question_recreation_reward_func": 0.39654287649318576,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.03125,
-      "rewards/xmlcount_reward_func": 0.9129062667489052,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.32137500308454037,
       "step": 14
     },
     {
-      "completion_length": 242.375,
-      "epoch": 4.0,
-      "grad_norm": 15.057260513305664,
-      "kl": 0.24116371832011887,
+      "completion_length": 381.40625,
+      "epoch": 0.64,
+      "grad_norm": 0.48140671849250793,
+      "kl": 0.0009308884182246402,
       "learning_rate": 8.067960709356478e-08,
-      "loss": 0.0002,
-      "reward": 2.993508279323578,
-      "reward_std": 0.7649101791903377,
-      "rewards/concensus_correctness_reward_func": 0.6969999919335047,
-      "rewards/consensus_reward_func": 1.0,
+      "loss": 0.0,
+      "reward": 0.6811805851757526,
+      "reward_std": 0.7240518294274807,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.48246658593416214,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.4305868395604193,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.8140416815876961,
+      "rewards/xmlcount_reward_func": 0.1880937460809946,
       "step": 16
     },
     {
-      "completion_length": 272.3125,
-      "epoch": 4.571428571428571,
-      "grad_norm": 7.568920612335205,
-      "kl": 0.015125435777008533,
+      "completion_length": 433.0,
+      "epoch": 0.72,
+      "grad_norm": 0.6308753490447998,
+      "kl": 0.0016986687787721166,
       "learning_rate": 3.013156219837776e-08,
       "loss": 0.0,
-      "reward": 2.294464409351349,
-      "reward_std": 0.6057919536251575,
-      "rewards/concensus_correctness_reward_func": 0.2955624982714653,
-      "rewards/consensus_reward_func": 0.5,
+      "reward": 0.8262888127937913,
+      "reward_std": 0.982679390348494,
+      "rewards/concensus_correctness_reward_func": 0.046875,
+      "rewards/consensus_reward_func": 0.0625,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.5682144520105794,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.40941380662843585,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.9150625020265579,
+      "rewards/xmlcount_reward_func": 0.22937500104308128,
       "step": 18
     },
     {
-      "completion_length": 286.2083333333333,
-      "epoch": 5.0,
-      "grad_norm": 7.497201442718506,
-      "kl": 0.05743966482502098,
+      "completion_length": 361.0625,
+      "epoch": 0.8,
+      "grad_norm": 0.5240185260772705,
+      "kl": 0.0009001127618830651,
       "learning_rate": 3.4096741493194193e-09,
       "loss": 0.0,
-      "reward": 2.476698577404022,
-      "reward_std": 1.642265244697531,
-      "rewards/concensus_correctness_reward_func": 0.4688333297769229,
-      "rewards/consensus_reward_func": 0.6666666666666666,
+      "reward": 0.9645268190652132,
+      "reward_std": 0.9338240548968315,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.1875,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.575615248332421,
+      "rewards/question_recreation_reward_func": 0.5564955612644553,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.765583356221517,
+      "rewards/strict_format_reward_func": 0.046875,
+      "rewards/xmlcount_reward_func": 0.17365624755620956,
       "step": 20
     },
     {
-      "epoch": 5.0,
+      "epoch": 0.8,
       "step": 20,
       "total_flos": 0.0,
-      "train_loss": 2.8217583894729616e-05,
-      "train_runtime": 239.7783,
-      "train_samples_per_second": 1.335,
-      "train_steps_per_second": 0.083
+      "train_loss": 9.213837813604187e-07,
+      "train_runtime": 1717.4399,
+      "train_samples_per_second": 0.186,
+      "train_steps_per_second": 0.012
     }
   ],
   "logging_steps": 2,
   "max_steps": 20,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 7,
+  "num_train_epochs": 1,
   "save_steps": 25,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -227,7 +227,7 @@
     }
   },
   "total_flos": 0.0,
-  "train_batch_size": 2,
+  "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3251af780e0893dd4521bf369e38701ef4125d1aeb4c2629271e51163d90dc04
-size 5880
+oid sha256:f9ba06b2ade5d86911b1b41a561f92e1e70c845897ea140bf725499a57c21075
+size 5944