Training in progress, step 400, checkpoint

checkpoint-400/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-400/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "codellama/CodeLlama-13b-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-400/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e3086c577bba91efe7a4205e7e957db94250eede3bf184e9ec3d98118007733
+size 26271757

checkpoint-400/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bcb1481a22b8dbfb3f000616906fe175796f01fe36c74556177b01a40555421
+size 13315525

checkpoint-400/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b93750c2e085667ff4452e14e3a0efee52b1c1ec542365d6fe8d095ff754ac3a
+size 14575

checkpoint-400/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0fd386e84f749a432dcea69ed33ac7fa38fb4997eff4d2d1dc66401045d18a3
+size 627

checkpoint-400/trainer_state.json

@@ -0,0 +1,105 @@
+{
+  "best_metric": 0.21197938919067383,
+  "best_model_checkpoint": "./text2sql/13b-codellama_instruct_pt_text2sql_ckpt8k_ft_spider/checkpoint-300",
+  "epoch": 1.4780600461893765,
+  "eval_steps": 100,
+  "global_step": 400,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.17283950617284e-07,
+      "loss": 0.5755,
+      "step": 1
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 3.08641975308642e-05,
+      "loss": 0.4793,
+      "step": 50
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.9251379038613084e-05,
+      "loss": 0.2673,
+      "step": 100
+    },
+    {
+      "epoch": 0.37,
+      "eval_loss": 0.25259310007095337,
+      "eval_runtime": 78.7093,
+      "eval_samples_per_second": 13.137,
+      "eval_steps_per_second": 3.291,
+      "step": 100
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.728132387706856e-05,
+      "loss": 0.2062,
+      "step": 150
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.5311268715524035e-05,
+      "loss": 0.1763,
+      "step": 200
+    },
+    {
+      "epoch": 0.74,
+      "eval_loss": 0.21567295491695404,
+      "eval_runtime": 78.7696,
+      "eval_samples_per_second": 13.127,
+      "eval_steps_per_second": 3.288,
+      "step": 200
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.334121355397952e-05,
+      "loss": 0.1569,
+      "step": 250
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.1371158392434986e-05,
+      "loss": 0.1387,
+      "step": 300
+    },
+    {
+      "epoch": 1.11,
+      "eval_loss": 0.21197938919067383,
+      "eval_runtime": 78.7959,
+      "eval_samples_per_second": 13.123,
+      "eval_steps_per_second": 3.287,
+      "step": 300
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 3.940110323089047e-05,
+      "loss": 0.131,
+      "step": 350
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 3.743104806934594e-05,
+      "loss": 0.125,
+      "step": 400
+    },
+    {
+      "epoch": 1.48,
+      "eval_loss": 0.21536056697368622,
+      "eval_runtime": 78.6623,
+      "eval_samples_per_second": 13.145,
+      "eval_steps_per_second": 3.293,
+      "step": 400
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 1350,
+  "num_train_epochs": 5,
+  "save_steps": 100,
+  "total_flos": 1.9592607420776448e+17,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-400/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54c7450b8be70b53ae3c8f7a3286d772311d477cf479dcc610962c6d0b68bfb8
+size 4283