Training in progress, step 22000, checkpoint

checkpoint-22000/README.md

@@ -0,0 +1,21 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- 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: bfloat16
+### Framework versions
+
+
+- PEFT 0.6.0.dev0

checkpoint-22000/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "codellama/CodeLlama-7b-Instruct-hf",
+  "bias": "lora_only",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 8,
+  "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-22000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57ac9a638012d9a59c281659d1ba638178b3647ffbee9f9f2bd845604f15ea8b
+size 16822989

checkpoint-22000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06230eaea1f9dae455c0d5f14858ee6835caeebdb23365ddaa62c0e71a68a8cf
+size 8555781

checkpoint-22000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b43766722ff911f35604fa71f01420c9e7a3b0e1e8cfd62bed055299e9dd782
+size 14575

checkpoint-22000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2dbd237b1e1d36dbcec04afa2f1cd75aa7edf4cf97dadf15d33cb45648441f0b
+size 627

checkpoint-22000/trainer_state.json

@@ -0,0 +1,245 @@
+{
+  "best_metric": 0.017868032678961754,
+  "best_model_checkpoint": "./text2sql/codellama_instruct_pt_text2sql/checkpoint-22000",
+  "epoch": 2.450741488546961,
+  "eval_steps": 2000,
+  "global_step": 22000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.1139992573338283e-07,
+      "loss": 1.2191,
+      "step": 1
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00011139992573338284,
+      "loss": 0.2684,
+      "step": 1000
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00022279985146676567,
+      "loss": 0.0693,
+      "step": 2000
+    },
+    {
+      "epoch": 0.22,
+      "eval_loss": 0.05887996032834053,
+      "eval_runtime": 171.127,
+      "eval_samples_per_second": 11.687,
+      "eval_steps_per_second": 1.461,
+      "step": 2000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00029781686301467276,
+      "loss": 0.0565,
+      "step": 3000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.0002907056676227274,
+      "loss": 0.047,
+      "step": 4000
+    },
+    {
+      "epoch": 0.45,
+      "eval_loss": 0.03957173973321915,
+      "eval_runtime": 171.3682,
+      "eval_samples_per_second": 11.671,
+      "eval_steps_per_second": 1.459,
+      "step": 4000
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00028359447223078195,
+      "loss": 0.0403,
+      "step": 5000
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.0002764832768388366,
+      "loss": 0.0364,
+      "step": 6000
+    },
+    {
+      "epoch": 0.67,
+      "eval_loss": 0.030724667012691498,
+      "eval_runtime": 170.9405,
+      "eval_samples_per_second": 11.7,
+      "eval_steps_per_second": 1.462,
+      "step": 6000
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0002693720814468912,
+      "loss": 0.0341,
+      "step": 7000
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00026226088605494583,
+      "loss": 0.0311,
+      "step": 8000
+    },
+    {
+      "epoch": 0.89,
+      "eval_loss": 0.027806701138615608,
+      "eval_runtime": 171.064,
+      "eval_samples_per_second": 11.692,
+      "eval_steps_per_second": 1.461,
+      "step": 8000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00025514969066300046,
+      "loss": 0.0289,
+      "step": 9000
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00024803849527105503,
+      "loss": 0.0251,
+      "step": 10000
+    },
+    {
+      "epoch": 1.11,
+      "eval_loss": 0.024082325398921967,
+      "eval_runtime": 171.4689,
+      "eval_samples_per_second": 11.664,
+      "eval_steps_per_second": 1.458,
+      "step": 10000
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00024092729987910966,
+      "loss": 0.0242,
+      "step": 11000
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00023381610448716428,
+      "loss": 0.0243,
+      "step": 12000
+    },
+    {
+      "epoch": 1.34,
+      "eval_loss": 0.02279273048043251,
+      "eval_runtime": 171.2147,
+      "eval_samples_per_second": 11.681,
+      "eval_steps_per_second": 1.46,
+      "step": 12000
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00022670490909521888,
+      "loss": 0.0219,
+      "step": 13000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.0002195937137032735,
+      "loss": 0.0227,
+      "step": 14000
+    },
+    {
+      "epoch": 1.56,
+      "eval_loss": 0.022252563387155533,
+      "eval_runtime": 171.1834,
+      "eval_samples_per_second": 11.683,
+      "eval_steps_per_second": 1.46,
+      "step": 14000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.00021248251831132813,
+      "loss": 0.0218,
+      "step": 15000
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 0.00020537132291938273,
+      "loss": 0.0212,
+      "step": 16000
+    },
+    {
+      "epoch": 1.78,
+      "eval_loss": 0.020056582987308502,
+      "eval_runtime": 170.9191,
+      "eval_samples_per_second": 11.701,
+      "eval_steps_per_second": 1.463,
+      "step": 16000
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 0.00019826012752743733,
+      "loss": 0.0208,
+      "step": 17000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 0.00019114893213549196,
+      "loss": 0.0202,
+      "step": 18000
+    },
+    {
+      "epoch": 2.01,
+      "eval_loss": 0.01821763999760151,
+      "eval_runtime": 171.4736,
+      "eval_samples_per_second": 11.664,
+      "eval_steps_per_second": 1.458,
+      "step": 18000
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.00018403773674354658,
+      "loss": 0.0161,
+      "step": 19000
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.0001769265413516012,
+      "loss": 0.016,
+      "step": 20000
+    },
+    {
+      "epoch": 2.23,
+      "eval_loss": 0.018352985382080078,
+      "eval_runtime": 171.2111,
+      "eval_samples_per_second": 11.681,
+      "eval_steps_per_second": 1.46,
+      "step": 20000
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 0.00016981534595965578,
+      "loss": 0.0155,
+      "step": 21000
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 0.0001627041505677104,
+      "loss": 0.0156,
+      "step": 22000
+    },
+    {
+      "epoch": 2.45,
+      "eval_loss": 0.017868032678961754,
+      "eval_runtime": 171.2047,
+      "eval_samples_per_second": 11.682,
+      "eval_steps_per_second": 1.46,
+      "step": 22000
+    }
+  ],
+  "logging_steps": 1000,
+  "max_steps": 44880,
+  "num_train_epochs": 5,
+  "save_steps": 2000,
+  "total_flos": 1.0708801647854813e+19,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-22000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c3233085912f12f47eb0b59979d9c767ddd5d0fc4cd96a7ebdcd9b12e9ba2c6
+size 4219