End of training

README.md

@@ -4,6 +4,7 @@ license: llama3.2
 base_model: meta-llama/Llama-3.2-3B
 tags:
 - llama-factory
+- lora
 - generated_from_trainer
 model-index:
 - name: lora_0-2_3B
@@ -15,7 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # lora_0-2_3B
 
-This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B) on the None dataset.
+This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B) on the gulaschnascher4000/stream-dataset-0-2, the identity-chatgulaschpt, the dolly_15k_de and the alpaca-gpt4_de datasets.
+It achieves the following results on the evaluation set:
+- Loss: 1.5461
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 0.10006807351940095,
+    "eval_loss": 1.5460742712020874,
+    "eval_runtime": 135.5079,
+    "eval_samples_per_second": 77.066,
+    "eval_steps_per_second": 19.268,
+    "total_flos": 8.37378777242665e+16,
+    "train_loss": 1.3456703529876917,
+    "train_runtime": 360.87,
+    "train_samples_per_second": 26.045,
+    "train_steps_per_second": 0.815
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 0.10006807351940095,
+    "eval_loss": 1.5460742712020874,
+    "eval_runtime": 135.5079,
+    "eval_samples_per_second": 77.066,
+    "eval_steps_per_second": 19.268
+}

runs/Jan14_01-43-47_pss5xry2f0aq/events.out.tfevents.1736819545.pss5xry2f0aq.48508.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7955d9b0bde15152a3d7eafd961a5b568ae441898cd860c4cc32fdc475e847e9
+size 359

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.10006807351940095,
+    "total_flos": 8.37378777242665e+16,
+    "train_loss": 1.3456703529876917,
+    "train_runtime": 360.87,
+    "train_samples_per_second": 26.045,
+    "train_steps_per_second": 0.815
+}

trainer_state.json

@@ -0,0 +1,245 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.10006807351940095,
+  "eval_steps": 500,
+  "global_step": 294,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0034036759700476512,
+      "grad_norm": 0.3913409113883972,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 1.517,
+      "step": 10
+    },
+    {
+      "epoch": 0.0068073519400953025,
+      "grad_norm": 0.42031803727149963,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 1.4607,
+      "step": 20
+    },
+    {
+      "epoch": 0.010211027910142955,
+      "grad_norm": 0.33206892013549805,
+      "learning_rate": 0.0001,
+      "loss": 1.3907,
+      "step": 30
+    },
+    {
+      "epoch": 0.013614703880190605,
+      "grad_norm": 0.38453030586242676,
+      "learning_rate": 9.964639423366442e-05,
+      "loss": 1.3338,
+      "step": 40
+    },
+    {
+      "epoch": 0.01701837985023826,
+      "grad_norm": 0.5517820715904236,
+      "learning_rate": 9.859057841617709e-05,
+      "loss": 1.3894,
+      "step": 50
+    },
+    {
+      "epoch": 0.02042205582028591,
+      "grad_norm": 0.3676927983760834,
+      "learning_rate": 9.68474862499881e-05,
+      "loss": 1.3543,
+      "step": 60
+    },
+    {
+      "epoch": 0.023825731790333562,
+      "grad_norm": 1.1669039726257324,
+      "learning_rate": 9.444177243274618e-05,
+      "loss": 1.3556,
+      "step": 70
+    },
+    {
+      "epoch": 0.02722940776038121,
+      "grad_norm": 0.4291529953479767,
+      "learning_rate": 9.140746393556854e-05,
+      "loss": 1.3505,
+      "step": 80
+    },
+    {
+      "epoch": 0.03063308373042886,
+      "grad_norm": 0.6486052870750427,
+      "learning_rate": 8.778747871771292e-05,
+      "loss": 1.3473,
+      "step": 90
+    },
+    {
+      "epoch": 0.03403675970047652,
+      "grad_norm": 0.42231065034866333,
+      "learning_rate": 8.363301868506264e-05,
+      "loss": 1.3506,
+      "step": 100
+    },
+    {
+      "epoch": 0.037440435670524165,
+      "grad_norm": 0.5026513934135437,
+      "learning_rate": 7.900284547855991e-05,
+      "loss": 1.3029,
+      "step": 110
+    },
+    {
+      "epoch": 0.04084411164057182,
+      "grad_norm": 0.6771671175956726,
+      "learning_rate": 7.396244933600285e-05,
+      "loss": 1.3336,
+      "step": 120
+    },
+    {
+      "epoch": 0.04424778761061947,
+      "grad_norm": 0.5867034792900085,
+      "learning_rate": 6.858312278301637e-05,
+      "loss": 1.3177,
+      "step": 130
+    },
+    {
+      "epoch": 0.047651463580667124,
+      "grad_norm": 0.5451478958129883,
+      "learning_rate": 6.294095225512603e-05,
+      "loss": 1.2924,
+      "step": 140
+    },
+    {
+      "epoch": 0.05105513955071477,
+      "grad_norm": 0.7218224406242371,
+      "learning_rate": 5.7115741913664264e-05,
+      "loss": 1.3566,
+      "step": 150
+    },
+    {
+      "epoch": 0.05445881552076242,
+      "grad_norm": 0.5916715264320374,
+      "learning_rate": 5.1189884877305375e-05,
+      "loss": 1.3093,
+      "step": 160
+    },
+    {
+      "epoch": 0.057862491490810075,
+      "grad_norm": 0.45692551136016846,
+      "learning_rate": 4.5247197834790876e-05,
+      "loss": 1.3207,
+      "step": 170
+    },
+    {
+      "epoch": 0.06126616746085772,
+      "grad_norm": 0.840484082698822,
+      "learning_rate": 3.937173552235117e-05,
+      "loss": 1.3437,
+      "step": 180
+    },
+    {
+      "epoch": 0.06466984343090537,
+      "grad_norm": 0.6113607883453369,
+      "learning_rate": 3.364660183412892e-05,
+      "loss": 1.3109,
+      "step": 190
+    },
+    {
+      "epoch": 0.06807351940095303,
+      "grad_norm": 0.5302855968475342,
+      "learning_rate": 2.8152774381532033e-05,
+      "loss": 1.308,
+      "step": 200
+    },
+    {
+      "epoch": 0.07147719537100068,
+      "grad_norm": 0.5346827507019043,
+      "learning_rate": 2.296795912722014e-05,
+      "loss": 1.331,
+      "step": 210
+    },
+    {
+      "epoch": 0.07488087134104833,
+      "grad_norm": 0.6379411816596985,
+      "learning_rate": 1.8165491294045593e-05,
+      "loss": 1.33,
+      "step": 220
+    },
+    {
+      "epoch": 0.07828454731109598,
+      "grad_norm": 0.4734562635421753,
+      "learning_rate": 1.3813298094746491e-05,
+      "loss": 1.3551,
+      "step": 230
+    },
+    {
+      "epoch": 0.08168822328114364,
+      "grad_norm": 0.680587887763977,
+      "learning_rate": 9.972937953781986e-06,
+      "loss": 1.3258,
+      "step": 240
+    },
+    {
+      "epoch": 0.08509189925119129,
+      "grad_norm": 0.5624794960021973,
+      "learning_rate": 6.698729810778065e-06,
+      "loss": 1.3268,
+      "step": 250
+    },
+    {
+      "epoch": 0.08849557522123894,
+      "grad_norm": 0.5939269065856934,
+      "learning_rate": 4.036984820916723e-06,
+      "loss": 1.3556,
+      "step": 260
+    },
+    {
+      "epoch": 0.09189925119128659,
+      "grad_norm": 0.6130989193916321,
+      "learning_rate": 2.0253513192751373e-06,
+      "loss": 1.2661,
+      "step": 270
+    },
+    {
+      "epoch": 0.09530292716133425,
+      "grad_norm": 0.8308888673782349,
+      "learning_rate": 6.922823140906753e-07,
+      "loss": 1.3755,
+      "step": 280
+    },
+    {
+      "epoch": 0.0987066031313819,
+      "grad_norm": 0.8799161911010742,
+      "learning_rate": 5.663304084960186e-08,
+      "loss": 1.3005,
+      "step": 290
+    },
+    {
+      "epoch": 0.10006807351940095,
+      "step": 294,
+      "total_flos": 8.37378777242665e+16,
+      "train_loss": 1.3456703529876917,
+      "train_runtime": 360.87,
+      "train_samples_per_second": 26.045,
+      "train_steps_per_second": 0.815
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 294,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 50000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.37378777242665e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}