Model save

README.md

@@ -0,0 +1,69 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B
+datasets:
+- generator
+library_name: peft
+license: llama3
+tags:
+- trl
+- sft
+- generated_from_trainer
+model-index:
+- name: llama3.1-8b-gpt4o_100k_classification-k
+  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. -->
+
+# llama3.1-8b-gpt4o_100k_classification-k
+
+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: 1.6323
+
+## 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: 0.0003
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- total_eval_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 1.0751        | 0.9958 | 118  | 1.6323          |
+
+
+### Framework versions
+
+- PEFT 0.13.1.dev0
+- Transformers 4.45.1
+- Pytorch 2.4.1+cu121
+- Datasets 3.0.1
+- Tokenizers 0.20.0

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9957805907172996,
+    "total_flos": 3.48383355932246e+17,
+    "train_loss": 1.2928365105289523,
+    "train_runtime": 1374.4457,
+    "train_samples": 92634,
+    "train_samples_per_second": 5.507,
+    "train_steps_per_second": 0.086
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9957805907172996,
+    "total_flos": 3.48383355932246e+17,
+    "train_loss": 1.2928365105289523,
+    "train_runtime": 1374.4457,
+    "train_samples": 92634,
+    "train_samples_per_second": 5.507,
+    "train_steps_per_second": 0.086
+}

trainer_state.json

@@ -0,0 +1,218 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9957805907172996,
+  "eval_steps": 500,
+  "global_step": 118,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008438818565400843,
+      "grad_norm": 1.0222958326339722,
+      "learning_rate": 2.4999999999999998e-05,
+      "loss": 2.86,
+      "step": 1
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 1.6936215162277222,
+      "learning_rate": 0.000125,
+      "loss": 2.8152,
+      "step": 5
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 2.3276407718658447,
+      "learning_rate": 0.00025,
+      "loss": 2.5324,
+      "step": 10
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 6.024274826049805,
+      "learning_rate": 0.0002994074763922825,
+      "loss": 1.692,
+      "step": 15
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.993145763874054,
+      "learning_rate": 0.0002958034371120616,
+      "loss": 1.4859,
+      "step": 20
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.7716729640960693,
+      "learning_rate": 0.000289003441114775,
+      "loss": 1.326,
+      "step": 25
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.5372955203056335,
+      "learning_rate": 0.0002791565417651033,
+      "loss": 1.2597,
+      "step": 30
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.5266535878181458,
+      "learning_rate": 0.00026647857940770634,
+      "loss": 1.1995,
+      "step": 35
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.3248014748096466,
+      "learning_rate": 0.0002512474502277316,
+      "loss": 1.1785,
+      "step": 40
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.22964197397232056,
+      "learning_rate": 0.00023379701487054785,
+      "loss": 1.1528,
+      "step": 45
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.2224261313676834,
+      "learning_rate": 0.00021450978034147806,
+      "loss": 1.1343,
+      "step": 50
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.22749578952789307,
+      "learning_rate": 0.00019380851559554636,
+      "loss": 1.1165,
+      "step": 55
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.2278250902891159,
+      "learning_rate": 0.00017214698460037218,
+      "loss": 1.1106,
+      "step": 60
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.22940242290496826,
+      "learning_rate": 0.00015,
+      "loss": 1.0946,
+      "step": 65
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.20619849860668182,
+      "learning_rate": 0.00012785301539962782,
+      "loss": 1.0952,
+      "step": 70
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "grad_norm": 0.19738492369651794,
+      "learning_rate": 0.00010619148440445364,
+      "loss": 1.0891,
+      "step": 75
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "grad_norm": 0.2077159434556961,
+      "learning_rate": 8.549021965852197e-05,
+      "loss": 1.0828,
+      "step": 80
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "grad_norm": 0.1939452439546585,
+      "learning_rate": 6.620298512945214e-05,
+      "loss": 1.0723,
+      "step": 85
+    },
+    {
+      "epoch": 0.759493670886076,
+      "grad_norm": 0.20105819404125214,
+      "learning_rate": 4.8752549772268444e-05,
+      "loss": 1.0617,
+      "step": 90
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "grad_norm": 0.2106432467699051,
+      "learning_rate": 3.352142059229365e-05,
+      "loss": 1.0647,
+      "step": 95
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "grad_norm": 0.19021108746528625,
+      "learning_rate": 2.0843458234896666e-05,
+      "loss": 1.0778,
+      "step": 100
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "grad_norm": 0.23987625539302826,
+      "learning_rate": 1.0996558885224993e-05,
+      "loss": 1.0761,
+      "step": 105
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "grad_norm": 0.1889180988073349,
+      "learning_rate": 4.1965628879383875e-06,
+      "loss": 1.0657,
+      "step": 110
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "grad_norm": 0.19509291648864746,
+      "learning_rate": 5.925236077174655e-07,
+      "loss": 1.0751,
+      "step": 115
+    },
+    {
+      "epoch": 0.9957805907172996,
+      "eval_loss": 1.6322884559631348,
+      "eval_runtime": 1.1519,
+      "eval_samples_per_second": 1.736,
+      "eval_steps_per_second": 0.868,
+      "step": 118
+    },
+    {
+      "epoch": 0.9957805907172996,
+      "step": 118,
+      "total_flos": 3.48383355932246e+17,
+      "train_loss": 1.2928365105289523,
+      "train_runtime": 1374.4457,
+      "train_samples_per_second": 5.507,
+      "train_steps_per_second": 0.086
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 118,
+  "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": 3.48383355932246e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}