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-mottled_foraging_boar
 tags:
@@ -8,13 +8,14 @@ tags:
 - grpo
 - gensyn
 - I am mottled foraging boar
+- unsloth
 - trl
 licence: license
 ---
 
 # Model Card for gensyn-checkpoints-mottled_foraging_boar
 
-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": [
+    "o_proj",
+    "up_proj",
+    "k_proj",
+    "q_proj",
+    "v_proj",
+    "down_proj",
+    "gate_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 sha256:6ae3781806028d6768418f7387951afcb5b45d46d11ed0998f06b2ef0dab5c3a
+size 73911112

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 3.586895763874054e-06,
-    "train_runtime": 310.4043,
-    "train_samples": 28,
-    "train_samples_per_second": 1.031,
-    "train_steps_per_second": 0.064
+    "train_loss": 8.256496755620901e-07,
+    "train_runtime": 792.9752,
+    "train_samples": 20,
+    "train_samples_per_second": 0.404,
+    "train_steps_per_second": 0.025
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 3.586895763874054e-06,
-    "train_runtime": 310.4043,
-    "train_samples": 28,
-    "train_samples_per_second": 1.031,
-    "train_steps_per_second": 0.064
+    "train_loss": 8.256496755620901e-07,
+    "train_runtime": 792.9752,
+    "train_samples": 20,
+    "train_samples_per_second": 0.404,
+    "train_steps_per_second": 0.025
 }

trainer_state.json

@@ -2,7 +2,7 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 5.0,
+  "epoch": 4.0,
   "eval_steps": 500,
   "global_step": 20,
   "is_hyper_param_search": false,
@@ -10,209 +10,209 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "completion_length": 435.4375,
-      "epoch": 0.5714285714285714,
-      "grad_norm": 3.8439865112304688,
-      "kl": 0.0,
+      "completion_length": 373.65625,
+      "epoch": 0.4,
+      "grad_norm": 0.5207684636116028,
+      "kl": 0.0007586075225844979,
       "learning_rate": 5e-07,
-      "loss": -0.0,
-      "reward": 0.6914588066283613,
-      "reward_std": 0.9236088133766316,
-      "rewards/concensus_correctness_reward_func": 0.05974999815225601,
-      "rewards/consensus_reward_func": 0.0625,
+      "loss": 0.0,
+      "reward": 1.5128968507051468,
+      "reward_std": 0.8223058478906751,
+      "rewards/concensus_correctness_reward_func": 0.16687500476837158,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0625,
-      "rewards/question_recreation_reward_func": 0.3949900634761434,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.5611155815422535,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.11171875428408384,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.45678125880658627,
       "step": 2
     },
     {
-      "completion_length": 333.1666666666667,
-      "epoch": 1.0,
-      "grad_norm": 4.06553316116333,
-      "kl": 0.0011363048397470266,
+      "completion_length": 474.1875,
+      "epoch": 0.8,
+      "grad_norm": 0.38550859689712524,
+      "kl": 0.0006429980239772703,
       "learning_rate": 4.864543104251586e-07,
       "loss": 0.0,
-      "reward": 1.343650996685028,
-      "reward_std": 0.9110305064047376,
-      "rewards/concensus_correctness_reward_func": 0.03800000001986822,
-      "rewards/consensus_reward_func": 0.08333333333333333,
+      "reward": 1.2827103175222874,
+      "reward_std": 1.0431885309517384,
+      "rewards/concensus_correctness_reward_func": 0.16493750177323818,
+      "rewards/consensus_reward_func": 0.25,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.16666666666666666,
-      "rewards/question_recreation_reward_func": 0.4201510362327099,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.5484915748238564,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.6354999976853529,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.2411562385968864,
       "step": 4
     },
     {
-      "completion_length": 349.0,
-      "epoch": 1.5714285714285714,
-      "grad_norm": 5.977542400360107,
-      "kl": 0.002151765987946419,
+      "completion_length": 338.53125,
+      "epoch": 1.2,
+      "grad_norm": 0.5113670229911804,
+      "kl": 0.0007692890430917032,
       "learning_rate": 4.472851273490984e-07,
       "loss": 0.0,
-      "reward": 2.060482904314995,
-      "reward_std": 2.559483935125172,
-      "rewards/concensus_correctness_reward_func": 0.8796250000596046,
+      "reward": 1.6476968228816986,
+      "reward_std": 1.20890873670578,
+      "rewards/concensus_correctness_reward_func": 0.14875000156462193,
       "rewards/consensus_reward_func": 0.25,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.3125,
-      "rewards/question_recreation_reward_func": 0.43748294562101364,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.6705092824995518,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.1808750033378601,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.43781252577900887,
       "step": 6
     },
     {
-      "completion_length": 349.0,
-      "epoch": 2.0,
-      "grad_norm": 3.4864773750305176,
-      "kl": 0.002737251779763028,
+      "completion_length": 451.65625,
+      "epoch": 1.6,
+      "grad_norm": 0.41855862736701965,
+      "kl": 0.0007109564830898307,
       "learning_rate": 3.867370395306068e-07,
       "loss": 0.0,
-      "reward": 1.5876267155011494,
-      "reward_std": 0.7928874848100046,
-      "rewards/concensus_correctness_reward_func": 0.03283333328242103,
-      "rewards/consensus_reward_func": 0.3333333333333333,
+      "reward": 1.3696312587708235,
+      "reward_std": 1.298433069139719,
+      "rewards/concensus_correctness_reward_func": 0.1850625015795231,
+      "rewards/consensus_reward_func": 0.125,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.4747934063586096,
+      "rewards/question_recreation_reward_func": 0.5645687524229288,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.49666666115323704,
+      "rewards/xmlcount_reward_func": 0.24500000476837158,
       "step": 8
     },
     {
-      "completion_length": 409.59375,
-      "epoch": 2.571428571428571,
-      "grad_norm": 4.636322021484375,
-      "kl": 0.006173002642754,
+      "completion_length": 311.9375,
+      "epoch": 2.0,
+      "grad_norm": 0.47906720638275146,
+      "kl": 0.0007747576528345235,
       "learning_rate": 3.1137137178519977e-07,
       "loss": 0.0,
-      "reward": 1.8668802455067635,
-      "reward_std": 1.6170106320641935,
-      "rewards/concensus_correctness_reward_func": 0.625,
-      "rewards/consensus_reward_func": 0.0,
+      "reward": 1.3798505514860153,
+      "reward_std": 0.7688908018171787,
+      "rewards/concensus_correctness_reward_func": 0.00918749999254942,
+      "rewards/consensus_reward_func": 0.25,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.46563018672168255,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.5008192812092602,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.5262500150129199,
+      "rewards/xmlcount_reward_func": 0.49484375305473804,
       "step": 10
     },
     {
-      "completion_length": 369.25,
-      "epoch": 3.0,
-      "grad_norm": 4.06023645401001,
-      "kl": 0.004249369774091368,
+      "completion_length": 362.1875,
+      "epoch": 2.4,
+      "grad_norm": 0.40098610520362854,
+      "kl": 0.000663719889416825,
       "learning_rate": 2.2935516363191693e-07,
       "loss": 0.0,
-      "reward": 1.6554374198118846,
-      "reward_std": 0.6809886172413826,
-      "rewards/concensus_correctness_reward_func": 0.07999999821186066,
-      "rewards/consensus_reward_func": 0.08333333333333333,
+      "reward": 2.0114393904805183,
+      "reward_std": 1.3165306523442268,
+      "rewards/concensus_correctness_reward_func": 0.035374999046325684,
+      "rewards/consensus_reward_func": 0.3125,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.16666666666666666,
-      "rewards/question_recreation_reward_func": 0.6594790716965994,
+      "rewards/final_correctness_reward_func": 0.5,
+      "rewards/question_recreation_reward_func": 0.6973769329488277,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.6659583300352097,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.4505625171586871,
       "step": 12
     },
     {
-      "completion_length": 345.34375,
-      "epoch": 3.571428571428571,
-      "grad_norm": 4.189980506896973,
-      "kl": 0.006222577969310805,
+      "completion_length": 389.96875,
+      "epoch": 2.8,
+      "grad_norm": 0.4310614764690399,
+      "kl": 0.0007022780264378525,
       "learning_rate": 1.4957614383675767e-07,
       "loss": 0.0,
-      "reward": 1.3473324589431286,
-      "reward_std": 0.679174258490093,
-      "rewards/concensus_correctness_reward_func": 0.0078125,
-      "rewards/consensus_reward_func": 0.0625,
+      "reward": 1.5623468831181526,
+      "reward_std": 1.2749357894062996,
+      "rewards/concensus_correctness_reward_func": 0.04987499862909317,
+      "rewards/consensus_reward_func": 0.1875,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.5202699541114271,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.5790344029664993,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.5067499973811209,
+      "rewards/strict_format_reward_func": 0.03125,
+      "rewards/xmlcount_reward_func": 0.527187486179173,
       "step": 14
     },
     {
-      "completion_length": 369.7083333333333,
-      "epoch": 4.0,
-      "grad_norm": 3.0566606521606445,
-      "kl": 0.006038210412953049,
+      "completion_length": 425.0625,
+      "epoch": 3.2,
+      "grad_norm": 0.5637480020523071,
+      "kl": 0.0007123533869162202,
       "learning_rate": 8.067960709356478e-08,
       "loss": 0.0,
-      "reward": 1.7356221427520115,
-      "reward_std": 0.8265478493024906,
-      "rewards/concensus_correctness_reward_func": 0.11958333399767677,
-      "rewards/consensus_reward_func": 0.25,
+      "reward": 1.349385041743517,
+      "reward_std": 0.8006823733448982,
+      "rewards/concensus_correctness_reward_func": 0.05487500037997961,
+      "rewards/consensus_reward_func": 0.125,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.511413821640114,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.7224787995219231,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.6046250065167745,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.4314062427729368,
       "step": 16
     },
     {
-      "completion_length": 389.375,
-      "epoch": 4.571428571428571,
-      "grad_norm": 16.731468200683594,
-      "kl": 0.006534892905619927,
+      "completion_length": 438.90625,
+      "epoch": 3.6,
+      "grad_norm": 0.44155919551849365,
+      "kl": 0.000691769972036127,
       "learning_rate": 3.013156219837776e-08,
       "loss": 0.0,
-      "reward": 1.3630528301000595,
-      "reward_std": 0.9979011313989758,
-      "rewards/concensus_correctness_reward_func": 0.11162499710917473,
-      "rewards/consensus_reward_func": 0.125,
+      "reward": 1.5265234410762787,
+      "reward_std": 1.8065023124217987,
+      "rewards/concensus_correctness_reward_func": 0.19037500116974115,
+      "rewards/consensus_reward_func": 0.3125,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0625,
-      "rewards/question_recreation_reward_func": 0.5975528075359762,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.5589609369635582,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.4663750035688281,
+      "rewards/strict_format_reward_func": 0.03125,
+      "rewards/xmlcount_reward_func": 0.1834374899044633,
       "step": 18
     },
     {
-      "completion_length": 407.8333333333333,
-      "epoch": 5.0,
-      "grad_norm": 5.084082126617432,
-      "kl": 0.005891711577229823,
+      "completion_length": 352.71875,
+      "epoch": 4.0,
+      "grad_norm": 0.6714590191841125,
+      "kl": 0.0019426245708018541,
       "learning_rate": 3.4096741493194193e-09,
       "loss": 0.0,
-      "reward": 0.8431620399157206,
-      "reward_std": 1.1959138717502356,
-      "rewards/concensus_correctness_reward_func": 0.037499999006589256,
-      "rewards/consensus_reward_func": 0.08333333333333333,
+      "reward": 1.3446000292897224,
+      "reward_std": 0.9004850089550018,
+      "rewards/concensus_correctness_reward_func": 0.15587500110268593,
+      "rewards/consensus_reward_func": 0.125,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.3333333333333333,
-      "rewards/question_recreation_reward_func": 0.37187035474926233,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.5265062879770994,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.017124998072783153,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.4590937588363886,
       "step": 20
     },
     {
-      "epoch": 5.0,
+      "epoch": 4.0,
       "step": 20,
       "total_flos": 0.0,
-      "train_loss": 3.586895763874054e-06,
-      "train_runtime": 310.4043,
-      "train_samples_per_second": 1.031,
-      "train_steps_per_second": 0.064
+      "train_loss": 8.256496755620901e-07,
+      "train_runtime": 792.9752,
+      "train_samples_per_second": 0.404,
+      "train_steps_per_second": 0.025
     }
   ],
   "logging_steps": 2,
   "max_steps": 20,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 7,
+  "num_train_epochs": 4,
   "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 sha256:0210fff6982d9ef9eb963d45aeaa136e2cd91b278164fb4d74e9ca995d1c3eaa
-size 5880
+oid sha256:b78fd2c9ec8a1c2edbb82869190c6808f833372a7d253d9cd342cc921cbd99e1
+size 5944