m-ric
/

OpenR1-SmolLM2-1.7B-Instruct-Agentic

@@ -1,11 +1,9 @@
 ---
 base_model: HuggingFaceTB/SmolLM2-1.7B-Instruct
-datasets: smolagents/training-traces
 library_name: transformers
 model_name: OpenR1-SmolLM2-1.7B-Instruct-Agentic
 tags:
 - generated_from_trainer
-- open-r1
 - trl
 - sft
 licence: license
@@ -13,7 +11,7 @@ licence: license
 # Model Card for OpenR1-SmolLM2-1.7B-Instruct-Agentic
-This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-1.7B-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B-Instruct) on the [smolagents/training-traces](https://huggingface.co/datasets/smolagents/training-traces) dataset.
 It has been trained using [TRL](https://github.com/huggingface/trl).
 ## Quick start
@@ -29,7 +27,7 @@ print(output["generated_text"])
 ## Training procedure
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/huggingface/huggingface/runs/y4n5o68t)
 This model was trained with SFT.

 ---
 base_model: HuggingFaceTB/SmolLM2-1.7B-Instruct
 library_name: transformers
 model_name: OpenR1-SmolLM2-1.7B-Instruct-Agentic
 tags:
 - generated_from_trainer
 - trl
 - sft
 licence: license
 # Model Card for OpenR1-SmolLM2-1.7B-Instruct-Agentic
+This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-1.7B-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B-Instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 ## Quick start
 ## Training procedure
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/huggingface/huggingface/runs/obusc7q3)
 This model was trained with SFT.

all_results.json CHANGED Viewed

@@ -1,8 +1,8 @@
 {
-    "total_flos": 30936649433088.0,
-    "train_loss": 0.0,
-    "train_runtime": 1.1169,
     "train_samples": 1204,
-    "train_samples_per_second": 1259.682,
-    "train_steps_per_second": 18.801
 }

 {
+    "total_flos": 6155761876992.0,
+    "train_loss": 1.0376410086949666,
+    "train_runtime": 169.1619,
     "train_samples": 1204,
+    "train_samples_per_second": 8.317,
+    "train_steps_per_second": 0.124
 }

train_results.json CHANGED Viewed

@@ -1,8 +1,8 @@
 {
-    "total_flos": 30936649433088.0,
-    "train_loss": 0.0,
-    "train_runtime": 1.1169,
     "train_samples": 1204,
-    "train_samples_per_second": 1259.682,
-    "train_steps_per_second": 18.801
 }

 {
+    "total_flos": 6155761876992.0,
+    "train_loss": 1.0376410086949666,
+    "train_runtime": 169.1619,
     "train_samples": 1204,
+    "train_samples_per_second": 8.317,
+    "train_steps_per_second": 0.124
 }

trainer_state.json CHANGED Viewed

@@ -2,201 +2,59 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 10.0,
   "eval_steps": 500,
-  "global_step": 100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.8,
-      "grad_norm": 0.45298852266312745,
-      "learning_rate": 1e-05,
-      "loss": 1.2979,
-      "mean_token_accuracy": 0.7309286197026571,
-      "num_tokens": 1946617.0,
-      "step": 5
-    },
-    {
-      "epoch": 1.0,
-      "grad_norm": 0.40654940219848923,
-      "learning_rate": 5e-05,
-      "loss": 1.2213,
-      "mean_token_accuracy": 0.7395898401737213,
-      "num_tokens": 2027951.0,
-      "step": 10
-    },
-    {
-      "epoch": 1.5,
-      "grad_norm": 0.2847470811451423,
-      "learning_rate": 4.722222222222222e-05,
-      "loss": 1.0874,
-      "mean_token_accuracy": 0.7629910290241242,
-      "num_tokens": 4554224.0,
-      "step": 15
-    },
-    {
-      "epoch": 2.0,
-      "grad_norm": 0.18223384267847015,
       "learning_rate": 4.4444444444444447e-05,
-      "loss": 0.9862,
-      "mean_token_accuracy": 0.7828312337398529,
-      "num_tokens": 7110068.0,
-      "step": 20
-    },
-    {
-      "epoch": 2.5,
-      "grad_norm": 0.14795383975219814,
-      "learning_rate": 4.166666666666667e-05,
-      "loss": 0.9164,
-      "mean_token_accuracy": 0.7957836747169494,
-      "num_tokens": 9668383.0,
-      "step": 25
-    },
-    {
-      "epoch": 3.0,
-      "grad_norm": 0.13696459830104152,
-      "learning_rate": 3.888888888888889e-05,
-      "loss": 0.8717,
-      "mean_token_accuracy": 0.8046842753887177,
-      "num_tokens": 12192185.0,
-      "step": 30
-    },
-    {
-      "epoch": 3.5,
-      "grad_norm": 0.11364616589196232,
-      "learning_rate": 3.611111111111111e-05,
-      "loss": 0.8405,
-      "mean_token_accuracy": 0.8104514122009278,
-      "num_tokens": 14727805.0,
-      "step": 35
-    },
-    {
-      "epoch": 4.0,
-      "grad_norm": 0.11796516723586992,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.7997,
-      "mean_token_accuracy": 0.8168916702270508,
-      "num_tokens": 17274302.0,
-      "step": 40
     },
     {
-      "epoch": 4.5,
-      "grad_norm": 0.11842645074111845,
       "learning_rate": 3.055555555555556e-05,
-      "loss": 0.7551,
-      "mean_token_accuracy": 0.8266302287578583,
-      "num_tokens": 19783780.0,
-      "step": 45
-    },
-    {
-      "epoch": 5.0,
-      "grad_norm": 0.10511642047034364,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.7741,
-      "mean_token_accuracy": 0.8223041534423828,
-      "num_tokens": 22356419.0,
-      "step": 50
-    },
-    {
-      "epoch": 5.5,
-      "grad_norm": 0.09869455025230396,
-      "learning_rate": 2.5e-05,
-      "loss": 0.7275,
-      "mean_token_accuracy": 0.8318577527999877,
-      "num_tokens": 24887270.0,
-      "step": 55
-    },
-    {
-      "epoch": 6.0,
-      "grad_norm": 0.10222368245808963,
-      "learning_rate": 2.2222222222222223e-05,
-      "loss": 0.7091,
-      "mean_token_accuracy": 0.835788244009018,
-      "num_tokens": 27438536.0,
-      "step": 60
-    },
-    {
-      "epoch": 6.5,
-      "grad_norm": 0.10330660110696606,
-      "learning_rate": 1.9444444444444445e-05,
-      "loss": 0.6933,
-      "mean_token_accuracy": 0.8389557778835297,
-      "num_tokens": 29966273.0,
-      "step": 65
     },
     {
-      "epoch": 7.0,
-      "grad_norm": 0.09938432484517981,
       "learning_rate": 1.6666666666666667e-05,
-      "loss": 0.6679,
-      "mean_token_accuracy": 0.8443207144737244,
-      "num_tokens": 32520653.0,
-      "step": 70
-    },
-    {
-      "epoch": 7.5,
-      "grad_norm": 0.10806152124483398,
-      "learning_rate": 1.388888888888889e-05,
-      "loss": 0.659,
-      "mean_token_accuracy": 0.8463097810745239,
-      "num_tokens": 35077955.0,
-      "step": 75
-    },
-    {
-      "epoch": 8.0,
-      "grad_norm": 0.10203259245493806,
-      "learning_rate": 1.1111111111111112e-05,
-      "loss": 0.6433,
-      "mean_token_accuracy": 0.8495967864990235,
-      "num_tokens": 37602770.0,
-      "step": 80
-    },
-    {
-      "epoch": 8.5,
-      "grad_norm": 0.10380131304938053,
-      "learning_rate": 8.333333333333334e-06,
-      "loss": 0.6315,
-      "mean_token_accuracy": 0.8518205523490906,
-      "num_tokens": 40123096.0,
-      "step": 85
-    },
-    {
-      "epoch": 9.0,
-      "grad_norm": 0.10787578796789908,
-      "learning_rate": 5.555555555555556e-06,
-      "loss": 0.6285,
-      "mean_token_accuracy": 0.8533558428287507,
-      "num_tokens": 42684887.0,
-      "step": 90
     },
     {
-      "epoch": 9.5,
-      "grad_norm": 0.10352907202385529,
       "learning_rate": 2.777777777777778e-06,
-      "loss": 0.6161,
-      "mean_token_accuracy": 0.8554697334766388,
-      "num_tokens": 45235842.0,
-      "step": 95
-    },
-    {
-      "epoch": 10.0,
-      "grad_norm": 0.10763838091671823,
-      "learning_rate": 0.0,
-      "loss": 0.6202,
-      "mean_token_accuracy": 0.8551790714263916,
-      "num_tokens": 47767004.0,
-      "step": 100
     },
     {
-      "epoch": 10.0,
-      "step": 100,
-      "total_flos": 30936649433088.0,
-      "train_loss": 0.0,
-      "train_runtime": 1.1169,
-      "train_samples_per_second": 1259.682,
-      "train_steps_per_second": 18.801
     }
   ],
   "logging_steps": 5,
@@ -216,7 +74,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 30936649433088.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
+  "epoch": 2.6666666666666665,
   "eval_steps": 500,
+  "global_step": 21,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.3659521589282829,
       "learning_rate": 4.4444444444444447e-05,
+      "loss": 1.1898,
+      "mean_token_accuracy": 0.749215167760849,
+      "num_tokens": 2476935.0,
+      "step": 5
     },
     {
+      "epoch": 1.2666666666666666,
+      "grad_norm": 0.31543356085932234,
       "learning_rate": 3.055555555555556e-05,
+      "loss": 1.0408,
+      "mean_token_accuracy": 0.773130026128557,
+      "num_tokens": 4779069.0,
+      "step": 10
     },
     {
+      "epoch": 1.9333333333333333,
+      "grad_norm": 0.23778923448881445,
       "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.9778,
+      "mean_token_accuracy": 0.7825889229774475,
+      "num_tokens": 7274168.0,
+      "step": 15
     },
     {
+      "epoch": 2.533333333333333,
+      "grad_norm": 0.19692051437026512,
       "learning_rate": 2.777777777777778e-06,
+      "loss": 0.9744,
+      "mean_token_accuracy": 0.7865516278478835,
+      "num_tokens": 9521856.0,
+      "step": 20
     },
     {
+      "epoch": 2.6666666666666665,
+      "mean_token_accuracy": 0.7911556959152222,
+      "num_tokens": 10020101.0,
+      "step": 21,
+      "total_flos": 6155761876992.0,
+      "train_loss": 1.0376410086949666,
+      "train_runtime": 169.1619,
+      "train_samples_per_second": 8.317,
+      "train_steps_per_second": 0.124
     }
   ],
   "logging_steps": 5,
       "attributes": {}
     }
   },
+  "total_flos": 6155761876992.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null