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-clawed_leaping_toucan
 tags:
@@ -8,13 +8,14 @@ tags:
 - grpo
 - gensyn
 - I am clawed leaping toucan
+- unsloth
 - trl
 licence: license
 ---
 
 # Model Card for gensyn-checkpoints-clawed_leaping_toucan
 
-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
@@ -39,8 +40,8 @@ This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing
 
 - TRL: 0.15.2
 - Transformers: 4.51.3
-- Pytorch: 2.5.1
-- Datasets: 3.5.0
+- Pytorch: 2.6.0
+- Datasets: 3.6.0
 - Tokenizers: 0.21.1
 
 ## Citations

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",
+    "down_proj",
+    "gate_proj",
+    "q_proj",
+    "up_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 sha256:480f21ec1837bd316152be3cb8b60e5d205ebe39f275db531e51d1adbed5894e
+size 73911112

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 7.946134473968414e-06,
-    "train_runtime": 198.3078,
-    "train_samples": 28,
-    "train_samples_per_second": 1.614,
-    "train_steps_per_second": 0.101
+    "train_loss": 7.248813339799653e-07,
+    "train_runtime": 1624.2404,
+    "train_samples": 98,
+    "train_samples_per_second": 0.197,
+    "train_steps_per_second": 0.012
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 7.946134473968414e-06,
-    "train_runtime": 198.3078,
-    "train_samples": 28,
-    "train_samples_per_second": 1.614,
-    "train_steps_per_second": 0.101
+    "train_loss": 7.248813339799653e-07,
+    "train_runtime": 1624.2404,
+    "train_samples": 98,
+    "train_samples_per_second": 0.197,
+    "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": 0.8163265306122449,
   "eval_steps": 500,
   "global_step": 20,
   "is_hyper_param_search": false,
@@ -10,209 +10,209 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "completion_length": 245.96875,
-      "epoch": 0.5714285714285714,
-      "grad_norm": 7.138059616088867,
+      "completion_length": 347.5625,
+      "epoch": 0.08163265306122448,
+      "grad_norm": 0.49709150195121765,
       "kl": 0.0,
       "learning_rate": 5e-07,
       "loss": 0.0,
-      "reward": 1.978086668998003,
-      "reward_std": 1.549998540780507,
-      "rewards/concensus_correctness_reward_func": 0.7274375036358833,
-      "rewards/consensus_reward_func": 0.125,
+      "reward": 1.5538663044571877,
+      "reward_std": 0.8957561552524567,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.4375,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.40799288742709905,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.4717725142836571,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.7176562547683716,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.44146874733269215,
       "step": 2
     },
     {
-      "completion_length": 346.8333333333333,
-      "epoch": 1.0,
-      "grad_norm": 5.335450649261475,
-      "kl": 0.0022581211815122515,
+      "completion_length": 392.34375,
+      "epoch": 0.16326530612244897,
+      "grad_norm": 1.0190365314483643,
+      "kl": 0.00082904764713021,
       "learning_rate": 4.864543104251586e-07,
       "loss": 0.0,
-      "reward": 1.0637057324250538,
-      "reward_std": 0.8543582057269911,
-      "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.0,
+      "reward": 1.1829268708825111,
+      "reward_std": 0.7207849770784378,
+      "rewards/concensus_correctness_reward_func": 0.0078125,
+      "rewards/consensus_reward_func": 0.0625,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.16666666666666666,
-      "rewards/question_recreation_reward_func": 0.462789058064421,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.460333114489913,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.43425000955661136,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.5116562619805336,
       "step": 4
     },
     {
-      "completion_length": 237.0625,
-      "epoch": 1.5714285714285714,
-      "grad_norm": 5.941693305969238,
-      "kl": 0.0033424893372284714,
+      "completion_length": 404.84375,
+      "epoch": 0.24489795918367346,
+      "grad_norm": 4.023301601409912,
+      "kl": 0.0009292965260101482,
       "learning_rate": 4.472851273490984e-07,
       "loss": 0.0,
-      "reward": 1.9579456727951765,
-      "reward_std": 1.5606304440298118,
-      "rewards/concensus_correctness_reward_func": 0.666812501847744,
-      "rewards/consensus_reward_func": 0.0625,
+      "reward": 1.3389800153672695,
+      "reward_std": 1.119007632136345,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.25,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.125,
-      "rewards/question_recreation_reward_func": 0.36688314378261566,
+      "rewards/question_recreation_reward_func": 0.6427613086998463,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.7367500066757202,
+      "rewards/strict_format_reward_func": 0.046875,
+      "rewards/xmlcount_reward_func": 0.27434375137090683,
       "step": 6
     },
     {
-      "completion_length": 219.70833333333334,
-      "epoch": 2.0,
-      "grad_norm": 3.439047336578369,
-      "kl": 0.005905011979242166,
+      "completion_length": 600.53125,
+      "epoch": 0.32653061224489793,
+      "grad_norm": 0.36614489555358887,
+      "kl": 0.0006038047395122703,
       "learning_rate": 3.867370395306068e-07,
       "loss": 0.0,
-      "reward": 1.7574667781591415,
-      "reward_std": 1.1081556143860023,
-      "rewards/concensus_correctness_reward_func": 0.17208333810170492,
-      "rewards/consensus_reward_func": 0.25,
+      "reward": 1.1000889372080564,
+      "reward_std": 1.182983409613371,
+      "rewards/concensus_correctness_reward_func": 0.00787500012665987,
+      "rewards/consensus_reward_func": 0.375,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.16666666666666666,
-      "rewards/question_recreation_reward_func": 0.354966773185879,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.7000889368355274,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.8137500037749609,
+      "rewards/xmlcount_reward_func": -0.045374996960163116,
       "step": 8
     },
     {
-      "completion_length": 174.46875,
-      "epoch": 2.571428571428571,
-      "grad_norm": 6.413847923278809,
-      "kl": 0.005604281301202718,
+      "completion_length": 436.1875,
+      "epoch": 0.40816326530612246,
+      "grad_norm": 0.6466080546379089,
+      "kl": 0.0008341637658304535,
       "learning_rate": 3.1137137178519977e-07,
       "loss": 0.0,
-      "reward": 1.3979325704276562,
-      "reward_std": 0.33654866518918425,
+      "reward": 1.4170492440462112,
+      "reward_std": 0.8611738085746765,
       "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.125,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.125,
-      "rewards/question_recreation_reward_func": 0.34633880454930477,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.4472992364317179,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.9265937581658363,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.51662502810359,
       "step": 10
     },
     {
-      "completion_length": 172.125,
-      "epoch": 3.0,
-      "grad_norm": 4.729302406311035,
-      "kl": 0.012371947659024348,
+      "completion_length": 449.71875,
+      "epoch": 0.4897959183673469,
+      "grad_norm": 0.5686748027801514,
+      "kl": 0.0008683982341608498,
       "learning_rate": 2.2935516363191693e-07,
       "loss": 0.0,
-      "reward": 2.351278011997541,
-      "reward_std": 1.1179534547651808,
-      "rewards/concensus_correctness_reward_func": 0.3329166720310847,
-      "rewards/consensus_reward_func": 0.4166666666666667,
+      "reward": 1.99321947991848,
+      "reward_std": 1.314138986170292,
+      "rewards/concensus_correctness_reward_func": 0.06006250157952309,
+      "rewards/consensus_reward_func": 0.3125,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.08333333333333333,
-      "rewards/question_recreation_reward_func": 0.5985279902815819,
+      "rewards/final_correctness_reward_func": 0.625,
+      "rewards/question_recreation_reward_func": 0.4730007923208177,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.9198333447178205,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.507031261920929,
       "step": 12
     },
     {
-      "completion_length": 154.5,
-      "epoch": 3.571428571428571,
-      "grad_norm": 6.839085578918457,
-      "kl": 0.02112107082211878,
+      "completion_length": 390.1875,
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.4010433554649353,
+      "kl": 0.0008200560114346445,
       "learning_rate": 1.4957614383675767e-07,
       "loss": 0.0,
-      "reward": 1.7222793698310852,
-      "reward_std": 0.7603381713852286,
-      "rewards/concensus_correctness_reward_func": 0.06843750178813934,
-      "rewards/consensus_reward_func": 0.125,
+      "reward": 0.9437324814498425,
+      "reward_std": 0.9435709491372108,
+      "rewards/concensus_correctness_reward_func": 0.03175000101327896,
+      "rewards/consensus_reward_func": 0.0625,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0625,
-      "rewards/question_recreation_reward_func": 0.4770293580368161,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.4078262336552143,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.9893125109374523,
+      "rewards/strict_format_reward_func": 0.046875,
+      "rewards/xmlcount_reward_func": 0.3947812579572201,
       "step": 14
     },
     {
-      "completion_length": 201.79166666666666,
-      "epoch": 4.0,
-      "grad_norm": 7.2178497314453125,
-      "kl": 0.01169851414548854,
+      "completion_length": 426.9375,
+      "epoch": 0.6530612244897959,
+      "grad_norm": 0.42668047547340393,
+      "kl": 0.0008046694783843122,
       "learning_rate": 8.067960709356478e-08,
       "loss": 0.0,
-      "reward": 1.5704542646805446,
-      "reward_std": 0.9336798458049694,
-      "rewards/concensus_correctness_reward_func": 0.16166667143503824,
-      "rewards/consensus_reward_func": 0.16666666666666666,
+      "reward": 1.422092966735363,
+      "reward_std": 1.0166446715593338,
+      "rewards/concensus_correctness_reward_func": 0.0521249994635582,
+      "rewards/consensus_reward_func": 0.25,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.16666666666666666,
-      "rewards/question_recreation_reward_func": 0.418662591992567,
-      "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.6567916870117188,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.43712421134114265,
+      "rewards/soft_format_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.046875,
+      "rewards/xmlcount_reward_func": 0.3703437428921461,
       "step": 16
     },
     {
-      "completion_length": 246.5625,
-      "epoch": 4.571428571428571,
-      "grad_norm": 5.255609512329102,
-      "kl": 0.012498048308771104,
+      "completion_length": 496.0625,
+      "epoch": 0.7346938775510204,
+      "grad_norm": 0.315399169921875,
+      "kl": 0.0007264475461852271,
       "learning_rate": 3.013156219837776e-08,
       "loss": 0.0,
-      "reward": 1.578137394040823,
-      "reward_std": 0.5315014050574973,
-      "rewards/concensus_correctness_reward_func": 0.0078125,
+      "reward": 1.3017337992787361,
+      "reward_std": 1.078663855791092,
+      "rewards/concensus_correctness_reward_func": 0.0028750000055879354,
       "rewards/consensus_reward_func": 0.0625,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.125,
-      "rewards/question_recreation_reward_func": 0.49882487556897104,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.6079837717115879,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.884000001475215,
+      "rewards/strict_format_reward_func": 0.046875,
+      "rewards/xmlcount_reward_func": 0.269000006839633,
       "step": 18
     },
     {
-      "completion_length": 170.33333333333334,
-      "epoch": 5.0,
-      "grad_norm": 7.401615619659424,
-      "kl": 0.01765511992077033,
+      "completion_length": 399.96875,
+      "epoch": 0.8163265306122449,
+      "grad_norm": 0.5829342007637024,
+      "kl": 0.0007923786251922138,
       "learning_rate": 3.4096741493194193e-09,
       "loss": 0.0,
-      "reward": 1.9397808909416199,
-      "reward_std": 1.1341329428056877,
-      "rewards/concensus_correctness_reward_func": 0.24175000687440237,
-      "rewards/consensus_reward_func": 0.16666666666666666,
+      "reward": 1.70121268928051,
+      "reward_std": 1.1589545384049416,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.5,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.16666666666666666,
-      "rewards/question_recreation_reward_func": 0.48361418799807626,
+      "rewards/final_correctness_reward_func": 0.375,
+      "rewards/question_recreation_reward_func": 0.7043689340353012,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.8810833394527435,
+      "rewards/xmlcount_reward_func": 0.12184375338256359,
       "step": 20
     },
     {
-      "epoch": 5.0,
+      "epoch": 0.8163265306122449,
       "step": 20,
       "total_flos": 0.0,
-      "train_loss": 7.946134473968414e-06,
-      "train_runtime": 198.3078,
-      "train_samples_per_second": 1.614,
-      "train_steps_per_second": 0.101
+      "train_loss": 7.248813339799653e-07,
+      "train_runtime": 1624.2404,
+      "train_samples_per_second": 0.197,
+      "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 sha256:4169b02347e043be4cff3fc21417547542f9e48ec63cde65aa905926046a4ebb
-size 5880
+oid sha256:58daae9439a5d747aa80b0e7e3abe0a9dbe1fc8cad663f4a5ef83f7d5bf22325
+size 5944