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-foraging_peckish_badger
 tags:
@@ -8,13 +8,14 @@ tags:
 - grpo
 - gensyn
 - I am foraging peckish badger
+- unsloth
 - trl
 licence: license
 ---
 
 # Model Card for gensyn-checkpoints-foraging_peckish_badger
 
-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",
+    "v_proj",
+    "k_proj",
+    "gate_proj",
+    "q_proj",
+    "up_proj",
+    "down_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 sha256:5d6be4da6a225c5853f48a2b05a842edd76c4992730b4939a11c785ceada5ef7
+size 73911112

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 2.2105872631072997e-06,
-    "train_runtime": 1082.617,
-    "train_samples": 28,
-    "train_samples_per_second": 0.296,
-    "train_steps_per_second": 0.018
+    "train_loss": 7.820151040505152e-07,
+    "train_runtime": 937.272,
+    "train_samples": 11,
+    "train_samples_per_second": 0.341,
+    "train_steps_per_second": 0.021
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 2.2105872631072997e-06,
-    "train_runtime": 1082.617,
-    "train_samples": 28,
-    "train_samples_per_second": 0.296,
-    "train_steps_per_second": 0.018
+    "train_loss": 7.820151040505152e-07,
+    "train_runtime": 937.272,
+    "train_samples": 11,
+    "train_samples_per_second": 0.341,
+    "train_steps_per_second": 0.021
 }

trainer_state.json

@@ -2,7 +2,7 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 5.0,
+  "epoch": 6.7272727272727275,
   "eval_steps": 500,
   "global_step": 20,
   "is_hyper_param_search": false,
@@ -10,209 +10,209 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "completion_length": 435.78125,
-      "epoch": 0.5714285714285714,
-      "grad_norm": 5.068814754486084,
-      "kl": 0.0,
+      "completion_length": 386.875,
+      "epoch": 0.7272727272727273,
+      "grad_norm": 0.648298978805542,
+      "kl": 0.0007997459979378618,
       "learning_rate": 5e-07,
-      "loss": -0.0,
-      "reward": 0.19804235943593085,
-      "reward_std": 0.8535868365433998,
+      "loss": 0.0,
+      "reward": 0.9969208752736449,
+      "reward_std": 0.7273423410952091,
       "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.0625,
+      "rewards/consensus_reward_func": 0.25,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.2692611172096804,
+      "rewards/question_recreation_reward_func": 0.5391396079212427,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": -0.1337187523022294,
+      "rewards/xmlcount_reward_func": 0.20778125058859587,
       "step": 2
     },
     {
-      "completion_length": 355.4583333333333,
-      "epoch": 1.0,
-      "grad_norm": 2.809861898422241,
-      "kl": 0.0009269307289893428,
+      "completion_length": 436.39285714285717,
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.6724653840065002,
+      "kl": 0.0007738028403504618,
       "learning_rate": 4.864543104251586e-07,
       "loss": 0.0,
-      "reward": 0.7621200631062189,
-      "reward_std": 0.6857940470799804,
-      "rewards/concensus_correctness_reward_func": 0.03125,
-      "rewards/consensus_reward_func": 0.0,
+      "reward": 1.0942734948226385,
+      "reward_std": 1.222027497632163,
+      "rewards/concensus_correctness_reward_func": 0.06107142980077437,
+      "rewards/consensus_reward_func": 0.2857142857142857,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.08333333333333333,
-      "rewards/question_recreation_reward_func": 0.3884950599943598,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.33166637803827015,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.2590416657427947,
+      "rewards/strict_format_reward_func": 0.03571428571428571,
+      "rewards/xmlcount_reward_func": 0.38010714309556143,
       "step": 4
     },
     {
-      "completion_length": 509.0625,
-      "epoch": 1.5714285714285714,
-      "grad_norm": 5.338972091674805,
-      "kl": 0.0013124001307005528,
+      "completion_length": 344.32142857142856,
+      "epoch": 2.0,
+      "grad_norm": 0.5057637691497803,
+      "kl": 0.0007551076040337128,
       "learning_rate": 4.472851273490984e-07,
       "loss": 0.0,
-      "reward": 0.5905632148496807,
-      "reward_std": 0.6783959390595555,
+      "reward": 1.660817095211574,
+      "reward_std": 0.9632722671542849,
       "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.0625,
+      "rewards/consensus_reward_func": 0.5714285714285714,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.38218822481576353,
+      "rewards/final_correctness_reward_func": 0.21428571428571427,
+      "rewards/question_recreation_reward_func": 0.5432099550962448,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.14587499666959047,
+      "rewards/xmlcount_reward_func": 0.3318928565297808,
       "step": 6
     },
     {
-      "completion_length": 464.9166666666667,
-      "epoch": 2.0,
-      "grad_norm": 2.9061334133148193,
-      "kl": 0.0018105933850165457,
+      "completion_length": 398.59375,
+      "epoch": 2.7272727272727275,
+      "grad_norm": 0.5353056192398071,
+      "kl": 0.0007648886203242,
       "learning_rate": 3.867370395306068e-07,
       "loss": 0.0,
-      "reward": 1.1275611594319344,
-      "reward_std": 0.5520922769792378,
-      "rewards/concensus_correctness_reward_func": 0.03566666692495346,
-      "rewards/consensus_reward_func": 0.16666666666666666,
+      "reward": 0.9478725977241993,
+      "reward_std": 0.8444725926965475,
+      "rewards/concensus_correctness_reward_func": 0.0018749999580904841,
+      "rewards/consensus_reward_func": 0.125,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.16666666666666666,
-      "rewards/question_recreation_reward_func": 0.4374361379692952,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.48312261048704386,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.32112500071525574,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.259750007186085,
       "step": 8
     },
     {
-      "completion_length": 443.6875,
-      "epoch": 2.571428571428571,
-      "grad_norm": 4.246496677398682,
-      "kl": 0.0023866116971476004,
+      "completion_length": 484.85714285714283,
+      "epoch": 3.3636363636363638,
+      "grad_norm": 0.27738702297210693,
+      "kl": 0.0006770965826165463,
       "learning_rate": 3.1137137178519977e-07,
       "loss": 0.0,
-      "reward": 0.539751447737217,
-      "reward_std": 0.8232741246465594,
-      "rewards/concensus_correctness_reward_func": 0.030249999836087227,
-      "rewards/consensus_reward_func": 0.0625,
+      "reward": 0.8648012195314679,
+      "reward_std": 1.1064413019589014,
+      "rewards/concensus_correctness_reward_func": 0.041214285152299066,
+      "rewards/consensus_reward_func": 0.35714285714285715,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.34272021567448974,
+      "rewards/question_recreation_reward_func": 0.34233695907252176,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.08865625702310354,
+      "rewards/strict_format_reward_func": 0.017857142857142856,
+      "rewards/xmlcount_reward_func": 0.10624998382159642,
       "step": 10
     },
     {
-      "completion_length": 354.7083333333333,
-      "epoch": 3.0,
-      "grad_norm": 5.332976818084717,
-      "kl": 0.0030408767488552257,
+      "completion_length": 329.35714285714283,
+      "epoch": 4.0,
+      "grad_norm": 0.3826559782028198,
+      "kl": 0.0006940006112147655,
       "learning_rate": 2.2935516363191693e-07,
       "loss": 0.0,
-      "reward": 0.91666238134106,
-      "reward_std": 0.602380842436105,
-      "rewards/concensus_correctness_reward_func": 0.15333333611488342,
-      "rewards/consensus_reward_func": 0.08333333333333333,
+      "reward": 1.6527051712785448,
+      "reward_std": 0.8527378227029528,
+      "rewards/concensus_correctness_reward_func": 0.04071428520338876,
+      "rewards/consensus_reward_func": 0.2857142857142857,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.32712074120839435,
+      "rewards/final_correctness_reward_func": 0.21428571428571427,
+      "rewards/question_recreation_reward_func": 0.6211337234292712,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.10287499862412612,
+      "rewards/xmlcount_reward_func": 0.4908571371010372,
       "step": 12
     },
     {
-      "completion_length": 505.3125,
-      "epoch": 3.571428571428571,
-      "grad_norm": 3.084214687347412,
-      "kl": 0.0029202045043348335,
+      "completion_length": 374.46875,
+      "epoch": 4.7272727272727275,
+      "grad_norm": 0.5625499486923218,
+      "kl": 0.000772401788708521,
       "learning_rate": 1.4957614383675767e-07,
       "loss": 0.0,
-      "reward": 0.5424395808950067,
-      "reward_std": 0.7874002328608185,
-      "rewards/concensus_correctness_reward_func": 0.0078125,
-      "rewards/consensus_reward_func": 0.0,
+      "reward": 1.1694097146391869,
+      "reward_std": 0.8063520211726427,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.5,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0625,
-      "rewards/question_recreation_reward_func": 0.3602208390366286,
-      "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.11190625419840217,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.4131596917286515,
+      "rewards/soft_format_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.03125,
+      "rewards/xmlcount_reward_func": 0.20937499403953552,
       "step": 14
     },
     {
-      "completion_length": 433.375,
-      "epoch": 4.0,
-      "grad_norm": 3.453960657119751,
-      "kl": 0.00320227052240322,
+      "completion_length": 388.89285714285717,
+      "epoch": 5.363636363636363,
+      "grad_norm": 0.5131890177726746,
+      "kl": 0.0006978282887887742,
       "learning_rate": 8.067960709356478e-08,
       "loss": 0.0,
-      "reward": 0.29166297893971205,
-      "reward_std": 0.4712665615370497,
-      "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.0,
+      "reward": 1.5317158954484122,
+      "reward_std": 0.7606781870126724,
+      "rewards/concensus_correctness_reward_func": 0.019642857568604604,
+      "rewards/consensus_reward_func": 0.5714285714285714,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.28437130618840456,
+      "rewards/final_correctness_reward_func": 0.21428571428571427,
+      "rewards/question_recreation_reward_func": 0.5711087456771305,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.0072916727513074875,
+      "rewards/xmlcount_reward_func": 0.1552500096814973,
       "step": 16
     },
     {
-      "completion_length": 384.21875,
-      "epoch": 4.571428571428571,
-      "grad_norm": 4.947248935699463,
-      "kl": 0.005076818481029477,
+      "completion_length": 369.35714285714283,
+      "epoch": 6.0,
+      "grad_norm": 16.381929397583008,
+      "kl": 0.0011041910287791065,
       "learning_rate": 3.013156219837776e-08,
       "loss": 0.0,
-      "reward": 0.7677737702615559,
-      "reward_std": 0.6437986879027449,
-      "rewards/concensus_correctness_reward_func": 0.018937500193715096,
-      "rewards/consensus_reward_func": 0.0625,
+      "reward": 0.9362205990723201,
+      "reward_std": 0.9797024769442422,
+      "rewards/concensus_correctness_reward_func": 0.041214285152299066,
+      "rewards/consensus_reward_func": 0.21428571428571427,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0625,
-      "rewards/question_recreation_reward_func": 0.3506175228394568,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.5888991824218205,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.27321874909102917,
+      "rewards/xmlcount_reward_func": 0.09182142785617284,
       "step": 18
     },
     {
-      "completion_length": 427.125,
-      "epoch": 5.0,
-      "grad_norm": 2.5313291549682617,
-      "kl": 0.005946058634435758,
+      "completion_length": 499.78125,
+      "epoch": 6.7272727272727275,
+      "grad_norm": 1.6484057903289795,
+      "kl": 0.001340657730906969,
       "learning_rate": 3.4096741493194193e-09,
       "loss": 0.0,
-      "reward": 0.733763511137416,
-      "reward_std": 0.30780596130837995,
+      "reward": 0.7778393775224686,
+      "reward_std": 0.9814196899533272,
       "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.1875,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.08333333333333333,
-      "rewards/question_recreation_reward_func": 0.30355515765647095,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.43299564626067877,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.34687499950329465,
+      "rewards/xmlcount_reward_func": -0.030156243592500687,
       "step": 20
     },
     {
-      "epoch": 5.0,
+      "epoch": 6.7272727272727275,
       "step": 20,
       "total_flos": 0.0,
-      "train_loss": 2.2105872631072997e-06,
-      "train_runtime": 1082.617,
-      "train_samples_per_second": 0.296,
-      "train_steps_per_second": 0.018
+      "train_loss": 7.820151040505152e-07,
+      "train_runtime": 937.272,
+      "train_samples_per_second": 0.341,
+      "train_steps_per_second": 0.021
     }
   ],
   "logging_steps": 2,
   "max_steps": 20,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 7,
+  "num_train_epochs": 10,
   "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 sha256:3a7044348908e734b8fa79e26a88bf7a2ae8986c187a4c087c49cce6ad5559ff
-size 5880
+oid sha256:dc867c7007b12b6e82917a571df9cdf12c2d3043e8b1b2b7e628f9dc4e2fecdf
+size 5944