Model save

README.md

@@ -0,0 +1,69 @@
+---
+library_name: peft
+license: llama3
+base_model: meta-llama/Meta-Llama-3-8B
+tags:
+- trl
+- sft
+- generated_from_trainer
+datasets:
+- generator
+model-index:
+- name: llama3-8b-kasa-closedqa-11-v1
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# llama3-8b-kasa-closedqa-11-v1
+
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.3773
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0002
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 256
+- total_eval_batch_size: 128
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 1.6912        | 0.9961 | 129  | 2.3773          |
+
+
+### Framework versions
+
+- PEFT 0.13.1.dev0
+- Transformers 4.46.2
+- Pytorch 2.5.1+cu124
+- Datasets 3.1.0
+- Tokenizers 0.20.3

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9961389961389961,
+    "total_flos": 7.617195409366057e+17,
+    "train_loss": 1.8217888118684753,
+    "train_runtime": 663.1517,
+    "train_samples": 111440,
+    "train_samples_per_second": 49.983,
+    "train_steps_per_second": 0.195
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9961389961389961,
+    "total_flos": 7.617195409366057e+17,
+    "train_loss": 1.8217888118684753,
+    "train_runtime": 663.1517,
+    "train_samples": 111440,
+    "train_samples_per_second": 49.983,
+    "train_steps_per_second": 0.195
+}

trainer_state.json

@@ -0,0 +1,232 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9961389961389961,
+  "eval_steps": 500,
+  "global_step": 129,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007722007722007722,
+      "grad_norm": 1.7816834449768066,
+      "learning_rate": 1.5384615384615387e-05,
+      "loss": 2.391,
+      "step": 1
+    },
+    {
+      "epoch": 0.03861003861003861,
+      "grad_norm": 1.4342066049575806,
+      "learning_rate": 7.692307692307693e-05,
+      "loss": 2.3898,
+      "step": 5
+    },
+    {
+      "epoch": 0.07722007722007722,
+      "grad_norm": 0.5361077189445496,
+      "learning_rate": 0.00015384615384615385,
+      "loss": 2.3243,
+      "step": 10
+    },
+    {
+      "epoch": 0.11583011583011583,
+      "grad_norm": 0.5296446681022644,
+      "learning_rate": 0.00019985334138511237,
+      "loss": 2.2164,
+      "step": 15
+    },
+    {
+      "epoch": 0.15444015444015444,
+      "grad_norm": 0.6367804408073425,
+      "learning_rate": 0.0001982083682742156,
+      "loss": 2.0755,
+      "step": 20
+    },
+    {
+      "epoch": 0.19305019305019305,
+      "grad_norm": 0.5406314134597778,
+      "learning_rate": 0.00019476531711828027,
+      "loss": 1.9492,
+      "step": 25
+    },
+    {
+      "epoch": 0.23166023166023167,
+      "grad_norm": 0.37518641352653503,
+      "learning_rate": 0.0001895872260758688,
+      "loss": 1.8895,
+      "step": 30
+    },
+    {
+      "epoch": 0.2702702702702703,
+      "grad_norm": 0.3098143935203552,
+      "learning_rate": 0.00018276889981568906,
+      "loss": 1.833,
+      "step": 35
+    },
+    {
+      "epoch": 0.3088803088803089,
+      "grad_norm": 0.2785213589668274,
+      "learning_rate": 0.00017443517375622704,
+      "loss": 1.7927,
+      "step": 40
+    },
+    {
+      "epoch": 0.3474903474903475,
+      "grad_norm": 0.23454847931861877,
+      "learning_rate": 0.00016473862847818277,
+      "loss": 1.784,
+      "step": 45
+    },
+    {
+      "epoch": 0.3861003861003861,
+      "grad_norm": 0.20828357338905334,
+      "learning_rate": 0.00015385679615609042,
+      "loss": 1.7591,
+      "step": 50
+    },
+    {
+      "epoch": 0.4247104247104247,
+      "grad_norm": 0.2269953042268753,
+      "learning_rate": 0.00014198891015602646,
+      "loss": 1.7541,
+      "step": 55
+    },
+    {
+      "epoch": 0.46332046332046334,
+      "grad_norm": 0.22065143287181854,
+      "learning_rate": 0.00012935225731039348,
+      "loss": 1.739,
+      "step": 60
+    },
+    {
+      "epoch": 0.5019305019305019,
+      "grad_norm": 0.2357364445924759,
+      "learning_rate": 0.0001161781996552765,
+      "loss": 1.7335,
+      "step": 65
+    },
+    {
+      "epoch": 0.5405405405405406,
+      "grad_norm": 0.26511350274086,
+      "learning_rate": 0.00010270793846761347,
+      "loss": 1.7297,
+      "step": 70
+    },
+    {
+      "epoch": 0.5791505791505791,
+      "grad_norm": 0.23851530253887177,
+      "learning_rate": 8.918809815760585e-05,
+      "loss": 1.7163,
+      "step": 75
+    },
+    {
+      "epoch": 0.6177606177606177,
+      "grad_norm": 0.273843914270401,
+      "learning_rate": 7.586621087002945e-05,
+      "loss": 1.7118,
+      "step": 80
+    },
+    {
+      "epoch": 0.6563706563706564,
+      "grad_norm": 0.2239556759595871,
+      "learning_rate": 6.298618446600856e-05,
+      "loss": 1.7166,
+      "step": 85
+    },
+    {
+      "epoch": 0.694980694980695,
+      "grad_norm": 0.242255300283432,
+      "learning_rate": 5.078383686109926e-05,
+      "loss": 1.6918,
+      "step": 90
+    },
+    {
+      "epoch": 0.7335907335907336,
+      "grad_norm": 0.21650227904319763,
+      "learning_rate": 3.948257848062351e-05,
+      "loss": 1.6965,
+      "step": 95
+    },
+    {
+      "epoch": 0.7722007722007722,
+      "grad_norm": 0.2475181221961975,
+      "learning_rate": 2.9289321881345254e-05,
+      "loss": 1.6979,
+      "step": 100
+    },
+    {
+      "epoch": 0.8108108108108109,
+      "grad_norm": 0.22755274176597595,
+      "learning_rate": 2.0390693429435627e-05,
+      "loss": 1.6902,
+      "step": 105
+    },
+    {
+      "epoch": 0.8494208494208494,
+      "grad_norm": 0.23448188602924347,
+      "learning_rate": 1.2949616394382802e-05,
+      "loss": 1.6914,
+      "step": 110
+    },
+    {
+      "epoch": 0.888030888030888,
+      "grad_norm": 0.25876742601394653,
+      "learning_rate": 7.102328018320858e-06,
+      "loss": 1.6983,
+      "step": 115
+    },
+    {
+      "epoch": 0.9266409266409267,
+      "grad_norm": 0.2385036200284958,
+      "learning_rate": 2.9558851746788517e-06,
+      "loss": 1.7022,
+      "step": 120
+    },
+    {
+      "epoch": 0.9652509652509652,
+      "grad_norm": 0.2816598117351532,
+      "learning_rate": 5.862042845640403e-07,
+      "loss": 1.6912,
+      "step": 125
+    },
+    {
+      "epoch": 0.9961389961389961,
+      "eval_loss": 2.3773248195648193,
+      "eval_runtime": 0.8215,
+      "eval_samples_per_second": 14.607,
+      "eval_steps_per_second": 1.217,
+      "step": 129
+    },
+    {
+      "epoch": 0.9961389961389961,
+      "step": 129,
+      "total_flos": 7.617195409366057e+17,
+      "train_loss": 1.8217888118684753,
+      "train_runtime": 663.1517,
+      "train_samples_per_second": 49.983,
+      "train_steps_per_second": 0.195
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 129,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 7.617195409366057e+17,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}