End of training

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: Gensyn/Qwen2.5-0.5B-Instruct
+base_model: unsloth/Qwen2.5-0.5B-Instruct
 library_name: transformers
 model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-fierce_padded_camel
 tags:
@@ -14,7 +14,7 @@ licence: license
 
 # Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-fierce_padded_camel
 
-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 [unsloth/Qwen2.5-0.5B-Instruct](https://huggingface.co/unsloth/Qwen2.5-0.5B-Instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 1.8783261111821047e-06,
-    "train_runtime": 35482.0828,
-    "train_samples": 17,
-    "train_samples_per_second": 0.005,
-    "train_steps_per_second": 0.0
+    "train_loss": 4.559755325317383e-07,
+    "train_runtime": 4512.8926,
+    "train_samples": 4,
+    "train_samples_per_second": 0.004,
+    "train_steps_per_second": 0.001
 }

config.json

@@ -3,7 +3,6 @@
     "Qwen2ForCausalLM"
   ],
   "attention_dropout": 0.0,
-  "bos_token_id": 151643,
   "eos_token_id": 151645,
   "hidden_act": "silu",
   "hidden_size": 896,
@@ -15,13 +14,15 @@
   "num_attention_heads": 14,
   "num_hidden_layers": 24,
   "num_key_value_heads": 2,
+  "pad_token_id": 151654,
   "rms_norm_eps": 1e-06,
   "rope_scaling": null,
   "rope_theta": 1000000.0,
-  "sliding_window": 32768,
+  "sliding_window": null,
   "tie_word_embeddings": true,
   "torch_dtype": "float32",
   "transformers_version": "4.51.2",
+  "unsloth_fixed": true,
   "use_cache": true,
   "use_sliding_window": false,
   "vocab_size": 151936

generation_config.json

@@ -5,7 +5,8 @@
     151645,
     151643
   ],
-  "pad_token_id": 151643,
+  "max_length": 32768,
+  "pad_token_id": 151654,
   "repetition_penalty": 1.1,
   "temperature": 0.7,
   "top_k": 20,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f1ead049fa7130a809557b15d02866f929a0da1bc89d3ae806bf16edb40e728
+oid sha256:c21dce77c4758024c73d72b3995fbb02834d57157e7fa5b7a1910e9e05a8754a
 size 1976163472

special_tokens_map.json

@@ -22,7 +22,7 @@
     "single_word": false
   },
   "pad_token": {
-    "content": "<|endoftext|>",
+    "content": "<|vision_pad|>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5eee858c5123a4279c3e1f7b81247343f356ac767940b2692a928ad929543214
-size 11422063
+oid sha256:64e71213db910f5cafa86d35091f37393dcc344b1bbc34091d1b3eed4cca01d5
+size 11422064

tokenizer_config.json

@@ -200,8 +200,9 @@
   "eos_token": "<|im_end|>",
   "errors": "replace",
   "extra_special_tokens": {},
-  "model_max_length": 131072,
-  "pad_token": "<|endoftext|>",
+  "model_max_length": 32768,
+  "pad_token": "<|vision_pad|>",
+  "padding_side": "left",
   "split_special_tokens": false,
   "tokenizer_class": "Qwen2Tokenizer",
   "unk_token": null

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 1.8783261111821047e-06,
-    "train_runtime": 35482.0828,
-    "train_samples": 17,
-    "train_samples_per_second": 0.005,
-    "train_steps_per_second": 0.0
+    "train_loss": 4.559755325317383e-07,
+    "train_runtime": 4512.8926,
+    "train_samples": 4,
+    "train_samples_per_second": 0.004,
+    "train_steps_per_second": 0.001
 }

trainer_state.json

@@ -2,123 +2,78 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.4705882352941178,
+  "epoch": 2.5,
   "eval_steps": 500,
-  "global_step": 10,
+  "global_step": 5,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "completion_length": 120.03125,
-      "epoch": 0.9411764705882353,
-      "grad_norm": 3.6002516746520996,
+      "completion_length": 116.875,
+      "epoch": 1.0,
+      "grad_norm": 14.425628662109375,
       "kl": 0.0,
       "learning_rate": 5e-07,
       "loss": -0.0,
-      "reward": 5.538293749094009,
-      "reward_std": 0.15910115039150696,
-      "rewards/concensus_correctness_reward_func": 1.922750011086464,
-      "rewards/consensus_reward_func": 2.0,
+      "reward": 0.3499542116187513,
+      "reward_std": 0.3656412720447406,
+      "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": 0.8246376209426671,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.09995422651991248,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.7909062542021275,
+      "rewards/xmlcount_reward_func": 0.0,
       "step": 2
     },
     {
-      "completion_length": 89.66666666666667,
-      "epoch": 1.4705882352941178,
-      "grad_norm": 4.7407546043396,
-      "kl": 0.0025498626831298075,
-      "learning_rate": 4.415111107797445e-07,
-      "loss": -0.0,
-      "reward": 5.3036660883161755,
-      "reward_std": 0.32777295475696316,
-      "rewards/concensus_correctness_reward_func": 1.8155555658870273,
-      "rewards/consensus_reward_func": 1.8888888888888888,
-      "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.7904995485312409,
-      "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.8087222245004442,
-      "step": 4
-    },
-    {
-      "completion_length": 130.5,
+      "completion_length": 127.875,
       "epoch": 2.0,
-      "grad_norm": 1.401399850845337,
-      "kl": 0.0020917902002111077,
-      "learning_rate": 2.934120444167326e-07,
-      "loss": -0.0,
-      "reward": 5.580838574303521,
-      "reward_std": 0.151048273831192,
-      "rewards/concensus_correctness_reward_func": 1.923555572827657,
-      "rewards/consensus_reward_func": 2.0,
-      "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.8693942957454257,
-      "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.7878888845443726,
-      "step": 6
-    },
-    {
-      "completion_length": 155.0625,
-      "epoch": 2.9411764705882355,
-      "grad_norm": 3.2692880630493164,
-      "kl": 0.005913253215112491,
-      "learning_rate": 1.2500000000000005e-07,
-      "loss": -0.0,
-      "reward": 5.509440541267395,
-      "reward_std": 0.2257549742462288,
-      "rewards/concensus_correctness_reward_func": 1.8626250103116035,
-      "rewards/consensus_reward_func": 1.9375,
+      "grad_norm": 26.35643768310547,
+      "kl": 0.0005819960133521818,
+      "learning_rate": 2.5e-07,
+      "loss": 0.0,
+      "reward": 0.11389446537941694,
+      "reward_std": 0.020214457530528307,
+      "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": 0.8912844185251743,
+      "rewards/question_recreation_reward_func": 0.11389446537941694,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.8180312439799309,
-      "step": 8
+      "rewards/xmlcount_reward_func": 0.0,
+      "step": 4
     },
     {
-      "completion_length": 112.66666666666667,
-      "epoch": 3.4705882352941178,
-      "grad_norm": 3.0253567695617676,
-      "kl": 0.007104916291104423,
-      "learning_rate": 1.507684480352292e-08,
-      "loss": 0.0,
-      "reward": 5.129255480236477,
-      "reward_std": 0.49882786147645675,
-      "rewards/concensus_correctness_reward_func": 1.8155555658870273,
-      "rewards/consensus_reward_func": 1.8888888888888888,
+      "completion_length": 78.75,
+      "epoch": 2.5,
+      "kl": 0.001228553446708247,
+      "reward": 0.06061847135424614,
+      "reward_std": 0.06433576252311468,
+      "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": 0.7083110366430547,
+      "rewards/question_recreation_reward_func": 0.06061847135424614,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.7164999975098504,
-      "step": 10
-    },
-    {
-      "epoch": 3.4705882352941178,
-      "step": 10,
+      "rewards/xmlcount_reward_func": 0.0,
+      "step": 5,
       "total_flos": 0.0,
-      "train_loss": 1.8783261111821047e-06,
-      "train_runtime": 35482.0828,
-      "train_samples_per_second": 0.005,
-      "train_steps_per_second": 0.0
+      "train_loss": 4.559755325317383e-07,
+      "train_runtime": 4512.8926,
+      "train_samples_per_second": 0.004,
+      "train_steps_per_second": 0.001
     }
   ],
   "logging_steps": 2,
-  "max_steps": 10,
+  "max_steps": 5,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
-  "save_steps": 25,
+  "num_train_epochs": 3,
+  "save_steps": 10,
   "stateful_callbacks": {
     "TrainerControl": {
       "args": {

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:02370b3c3295ea10eee42efa99a1972bbd5a3051d24268f26737720e89cf49f3
+oid sha256:0a6c138741c081665c9685f74471e67833840ee9cd96797f78c1171c1cf29e0e
 size 5944