commit from root

README.md

@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
+### Framework versions
+
+
+- PEFT 0.4.0

adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "./Llama-2-7b-chat-hf",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ea7b0fb91f08aca8b4b7ab4fc593a72c7ac8f936a8b1b0f74170dc31085ccd8
+size 16821197

all_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.0,
+    "train_loss": 0.7399377094979515,
+    "train_runtime": 176487.3633,
+    "train_samples_per_second": 2.562,
+    "train_steps_per_second": 0.16
+}

checkpoint-1000/README.md

@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
+### Framework versions
+
+
+- PEFT 0.4.0

checkpoint-1000/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "./Llama-2-7b-chat-hf",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-1000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a9a2d59099a538613c48a0b00ee727732ab2529a774194cb4fc25234c00c876
+size 16821197

checkpoint-1000/finetuning_args.json

@@ -0,0 +1,16 @@
+{
+  "dpo_beta": 0.1,
+  "finetuning_type": "lora",
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "lora_rank": 8,
+  "lora_target": [
+    "q_proj",
+    "v_proj"
+  ],
+  "name_module_trainable": "mlp",
+  "num_hidden_layers": 32,
+  "num_layer_trainable": 3,
+  "ppo_score_norm": false,
+  "resume_lora_training": true
+}

checkpoint-1000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa7f1c6570f9391d5d5017e923d9263b86cd14f345e7de3dcc6457dde24209e5
+size 33661637

checkpoint-1000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd0a027fab7d1d27eeff972e975255506e62b2efd84dd6b01dc103696d8f4e46
+size 18663

checkpoint-1000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4e98d8a4e42439109d00bd301cb15fbea0119270f18a90ec80ba6f284c47478
+size 627

checkpoint-1000/trainer_state.json

@@ -0,0 +1,616 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.07078267947833165,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999998454568244e-05,
+      "loss": 1.3539,
+      "step": 10
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9999938182748876e-05,
+      "loss": 1.1833,
+      "step": 20
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9999870029288556e-05,
+      "loss": 1.173,
+      "step": 30
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999976494017406e-05,
+      "loss": 1.0772,
+      "step": 40
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999962894271507e-05,
+      "loss": 1.0715,
+      "step": 50
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9999462037079705e-05,
+      "loss": 1.0268,
+      "step": 60
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999926422347434e-05,
+      "loss": 0.9807,
+      "step": 70
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999903550214352e-05,
+      "loss": 0.9862,
+      "step": 80
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999877587337004e-05,
+      "loss": 0.9725,
+      "step": 90
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999848533747488e-05,
+      "loss": 0.9993,
+      "step": 100
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999816389481725e-05,
+      "loss": 0.9596,
+      "step": 110
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999781154579456e-05,
+      "loss": 0.979,
+      "step": 120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9997428290842444e-05,
+      "loss": 0.9748,
+      "step": 130
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999701413043471e-05,
+      "loss": 0.9309,
+      "step": 140
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999656906508344e-05,
+      "loss": 0.9143,
+      "step": 150
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999609309533887e-05,
+      "loss": 0.9439,
+      "step": 160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999558622178947e-05,
+      "loss": 0.9286,
+      "step": 170
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.99950484450619e-05,
+      "loss": 0.9544,
+      "step": 180
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999447976582104e-05,
+      "loss": 0.9355,
+      "step": 190
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999388018476998e-05,
+      "loss": 0.9154,
+      "step": 200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999324970265001e-05,
+      "loss": 0.9326,
+      "step": 210
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.999258832024061e-05,
+      "loss": 0.9215,
+      "step": 220
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.99918960383595e-05,
+      "loss": 0.9281,
+      "step": 230
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9991172857862555e-05,
+      "loss": 0.935,
+      "step": 240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.99904187796439e-05,
+      "loss": 0.941,
+      "step": 250
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9989633804635814e-05,
+      "loss": 0.9377,
+      "step": 260
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9988817933808814e-05,
+      "loss": 0.9014,
+      "step": 270
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9987971168171585e-05,
+      "loss": 0.9323,
+      "step": 280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.998709350877103e-05,
+      "loss": 0.8987,
+      "step": 290
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.998618495669224e-05,
+      "loss": 0.8933,
+      "step": 300
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9985245513058495e-05,
+      "loss": 0.893,
+      "step": 310
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9984275179031276e-05,
+      "loss": 0.909,
+      "step": 320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.998327395581025e-05,
+      "loss": 0.9235,
+      "step": 330
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9982241844633265e-05,
+      "loss": 0.8945,
+      "step": 340
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.998117884677638e-05,
+      "loss": 0.9095,
+      "step": 350
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.998008496355382e-05,
+      "loss": 0.8919,
+      "step": 360
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9978960196318006e-05,
+      "loss": 0.9088,
+      "step": 370
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.997780454645954e-05,
+      "loss": 0.8985,
+      "step": 380
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.99766180154072e-05,
+      "loss": 0.8972,
+      "step": 390
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9975400604627957e-05,
+      "loss": 0.8983,
+      "step": 400
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9974152315626935e-05,
+      "loss": 0.9115,
+      "step": 410
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.997287314994746e-05,
+      "loss": 0.8957,
+      "step": 420
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.997156310917103e-05,
+      "loss": 0.8681,
+      "step": 430
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9970222194917296e-05,
+      "loss": 0.894,
+      "step": 440
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.996885040884409e-05,
+      "loss": 0.8798,
+      "step": 450
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.996744775264743e-05,
+      "loss": 0.9034,
+      "step": 460
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.996601422806147e-05,
+      "loss": 0.9033,
+      "step": 470
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9964549836858536e-05,
+      "loss": 0.8841,
+      "step": 480
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9963054580849134e-05,
+      "loss": 0.8877,
+      "step": 490
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.996152846188191e-05,
+      "loss": 0.8729,
+      "step": 500
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.995997148184369e-05,
+      "loss": 0.8853,
+      "step": 510
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9958383642659414e-05,
+      "loss": 0.8837,
+      "step": 520
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.995676494629221e-05,
+      "loss": 0.8833,
+      "step": 530
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9955115394743354e-05,
+      "loss": 0.8843,
+      "step": 540
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.995343499005225e-05,
+      "loss": 0.892,
+      "step": 550
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.995172373429646e-05,
+      "loss": 0.8575,
+      "step": 560
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9949981629591705e-05,
+      "loss": 0.8311,
+      "step": 570
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.99482086780918e-05,
+      "loss": 0.8669,
+      "step": 580
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.994640488198874e-05,
+      "loss": 0.8388,
+      "step": 590
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.994457024351264e-05,
+      "loss": 0.8424,
+      "step": 600
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.994270476493175e-05,
+      "loss": 0.8676,
+      "step": 610
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.994080844855243e-05,
+      "loss": 0.8598,
+      "step": 620
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.993888129671921e-05,
+      "loss": 0.824,
+      "step": 630
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.993692331181469e-05,
+      "loss": 0.8652,
+      "step": 640
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.993493449625963e-05,
+      "loss": 0.8533,
+      "step": 650
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.993291485251288e-05,
+      "loss": 0.8677,
+      "step": 660
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.993086438307143e-05,
+      "loss": 0.8459,
+      "step": 670
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9928783090470365e-05,
+      "loss": 0.8626,
+      "step": 680
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.992667097728287e-05,
+      "loss": 0.8127,
+      "step": 690
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.992452804612027e-05,
+      "loss": 0.8716,
+      "step": 700
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.992235429963195e-05,
+      "loss": 0.8544,
+      "step": 710
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.992014974050542e-05,
+      "loss": 0.8562,
+      "step": 720
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.991791437146627e-05,
+      "loss": 0.871,
+      "step": 730
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9915648195278186e-05,
+      "loss": 0.8453,
+      "step": 740
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9913351214742945e-05,
+      "loss": 0.8524,
+      "step": 750
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.991102343270042e-05,
+      "loss": 0.8581,
+      "step": 760
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9908664852028545e-05,
+      "loss": 0.8477,
+      "step": 770
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.990627547564335e-05,
+      "loss": 0.8651,
+      "step": 780
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.990385530649891e-05,
+      "loss": 0.8453,
+      "step": 790
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9901404347587404e-05,
+      "loss": 0.8586,
+      "step": 800
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9898922601939056e-05,
+      "loss": 0.8746,
+      "step": 810
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.989641007262218e-05,
+      "loss": 0.8652,
+      "step": 820
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.98938667627431e-05,
+      "loss": 0.8531,
+      "step": 830
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.989129267544626e-05,
+      "loss": 0.8686,
+      "step": 840
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.988868781391408e-05,
+      "loss": 0.8692,
+      "step": 850
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.988605218136711e-05,
+      "loss": 0.8274,
+      "step": 860
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9883385781063876e-05,
+      "loss": 0.8502,
+      "step": 870
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9880688616300975e-05,
+      "loss": 0.8445,
+      "step": 880
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9877960690413035e-05,
+      "loss": 0.8475,
+      "step": 890
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.987520200677271e-05,
+      "loss": 0.8215,
+      "step": 900
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.987241256879071e-05,
+      "loss": 0.8389,
+      "step": 910
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.986959237991571e-05,
+      "loss": 0.8422,
+      "step": 920
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.9866741443634455e-05,
+      "loss": 0.8287,
+      "step": 930
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.986385976347169e-05,
+      "loss": 0.8694,
+      "step": 940
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.986094734299016e-05,
+      "loss": 0.847,
+      "step": 950
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.985800418579063e-05,
+      "loss": 0.8191,
+      "step": 960
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.985503029551184e-05,
+      "loss": 0.8419,
+      "step": 970
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.985202567583057e-05,
+      "loss": 0.8517,
+      "step": 980
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.984899033046155e-05,
+      "loss": 0.8653,
+      "step": 990
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.9845924263157526e-05,
+      "loss": 0.8349,
+      "step": 1000
+    }
+  ],
+  "max_steps": 28254,
+  "num_train_epochs": 2,
+  "total_flos": 2.4351950380204032e+17,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:221face861d281c49061d94e69a5df2e8356d17457f5f4ef2f014d70fd21249c
+size 3271

checkpoint-10000/README.md

@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
+### Framework versions
+
+
+- PEFT 0.4.0

checkpoint-10000/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "./Llama-2-7b-chat-hf",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-10000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3f714c57ae0e84c325da10732bbcdc8c43ef4e2ea454b9bfe7595bef647318c
+size 16821197

checkpoint-10000/finetuning_args.json

@@ -0,0 +1,16 @@
+{
+  "dpo_beta": 0.1,
+  "finetuning_type": "lora",
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "lora_rank": 8,
+  "lora_target": [
+    "q_proj",
+    "v_proj"
+  ],
+  "name_module_trainable": "mlp",
+  "num_hidden_layers": 32,
+  "num_layer_trainable": 3,
+  "ppo_score_norm": false,
+  "resume_lora_training": true
+}

checkpoint-10000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42123fa83d920bb3a255a7a687a2eb1ea9f17e566bb97bb150b469a78b965ce1
+size 33661637

checkpoint-10000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a05d6225789398771b5189036c35cf28db936606ca14f38f8ecd1039e1dcde3a
+size 18663

checkpoint-10000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93cc76c5d7ae5ed46080b681d1f5208fe8bc5915697b321ef9f75faa596c26c9
+size 627

checkpoint-10000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:221face861d281c49061d94e69a5df2e8356d17457f5f4ef2f014d70fd21249c
+size 3271

checkpoint-11000/README.md

@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
+### Framework versions
+
+
+- PEFT 0.4.0

checkpoint-11000/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "./Llama-2-7b-chat-hf",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-11000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a26437afd89e8830252fa31d98f1f93d178e45eb47dec9daab089eda6cc14cb7
+size 16821197

checkpoint-11000/finetuning_args.json

@@ -0,0 +1,16 @@
+{
+  "dpo_beta": 0.1,
+  "finetuning_type": "lora",
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "lora_rank": 8,
+  "lora_target": [
+    "q_proj",
+    "v_proj"
+  ],
+  "name_module_trainable": "mlp",
+  "num_hidden_layers": 32,
+  "num_layer_trainable": 3,
+  "ppo_score_norm": false,
+  "resume_lora_training": true
+}

checkpoint-11000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3dfdc5a7b72a82bea6b1e623b952a95acc24b2e6a929041a0240338ae0d4a344
+size 33661637

checkpoint-11000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6109f33969d76a77ca5a4ec7325a2b0f5c2f0162418caf5905fe6064ad3e6074
+size 18663

checkpoint-11000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0eb3b6455e8be51eec964213c3e96c486647705885fc26f5616552411cb5375e
+size 627

checkpoint-11000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:221face861d281c49061d94e69a5df2e8356d17457f5f4ef2f014d70fd21249c
+size 3271

checkpoint-12000/README.md

@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
+### Framework versions
+
+
+- PEFT 0.4.0

checkpoint-12000/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "./Llama-2-7b-chat-hf",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-12000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15cdefa05b6a05bcd8506fe3040c564100608491ad329a6d0457fb705598efd1
+size 16821197

checkpoint-12000/finetuning_args.json

@@ -0,0 +1,16 @@
+{
+  "dpo_beta": 0.1,
+  "finetuning_type": "lora",
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "lora_rank": 8,
+  "lora_target": [
+    "q_proj",
+    "v_proj"
+  ],
+  "name_module_trainable": "mlp",
+  "num_hidden_layers": 32,
+  "num_layer_trainable": 3,
+  "ppo_score_norm": false,
+  "resume_lora_training": true
+}

checkpoint-12000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e54b9abfd43b35a4193c6e08f54701755cf5f80021e890fe8c9b4062c584767
+size 33661637

checkpoint-12000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:497fc5f4f84ce21c7b29d94b52ec8313d6cb1ed9a8c4299da58f29e636f1e066
+size 18663

checkpoint-12000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7ef392eb88915ec01d1b6e6bc31e16cd878df9a85102b1a2433fb0bdd4d5ee0
+size 627

checkpoint-12000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:221face861d281c49061d94e69a5df2e8356d17457f5f4ef2f014d70fd21249c
+size 3271

checkpoint-13000/README.md

@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
+### Framework versions
+
+
+- PEFT 0.4.0

checkpoint-13000/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "./Llama-2-7b-chat-hf",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-13000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38b36ab9ab0afc9ef5746482ede51e116a4ed0f4afd488519d80af3ff546f112
+size 16821197

checkpoint-13000/finetuning_args.json

@@ -0,0 +1,16 @@
+{
+  "dpo_beta": 0.1,
+  "finetuning_type": "lora",
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "lora_rank": 8,
+  "lora_target": [
+    "q_proj",
+    "v_proj"
+  ],
+  "name_module_trainable": "mlp",
+  "num_hidden_layers": 32,
+  "num_layer_trainable": 3,
+  "ppo_score_norm": false,
+  "resume_lora_training": true
+}

checkpoint-13000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4039ba5883d9c3a3811e634bdbf861fa23147fc2d3373da1603eef651eff800
+size 33661637

checkpoint-13000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2bd8d759d0ef06b2ecb227efae8b257b736c638ac6508a3810732b04842cf81f
+size 18663

checkpoint-13000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac65d947299795de87194e2e527c46bcce31fcd02f89111f6e6ba1050b5bf93b
+size 627

checkpoint-13000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:221face861d281c49061d94e69a5df2e8356d17457f5f4ef2f014d70fd21249c
+size 3271

checkpoint-14000/README.md

@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
+### Framework versions
+
+
+- PEFT 0.4.0