Upload folder using huggingface_hub

README.md

@@ -0,0 +1,23 @@
+---
+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
+- load_in_4bit: True
+- load_in_8bit: False
+### Framework versions
+
+
+- PEFT 0.4.0

adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 64,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "k_proj",
+    "o_proj",
+    "q_proj",
+    "v_proj",
+    "up_proj",
+    "down_proj"
+  ],
+  "task_type": "SEQ_CLS"
+}

adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9d5aaae4f76e991c745d0625feb8399589f5fed6ffa655ff1888c189f450b1b
+size 160102926

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.9,
+    "total_flos": 5.046529217789952e+16,
+    "train_loss": 0.7722815573215485,
+    "train_runtime": 16370.7549,
+    "train_samples_per_second": 0.157,
+    "train_steps_per_second": 0.002
+}

runs/Apr06_04-02-20_fa1115e4f8e3/events.out.tfevents.1712376142.fa1115e4f8e3.35.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a266fe737f01b10b55426464e3b4f50c7456f292322940576b1b5ebe6dbf93d7
+size 9283

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.9,
+    "total_flos": 5.046529217789952e+16,
+    "train_loss": 0.7722815573215485,
+    "train_runtime": 16370.7549,
+    "train_samples_per_second": 0.157,
+    "train_steps_per_second": 0.002
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.902439024390244,
+  "eval_steps": 8,
+  "global_step": 40,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.78,
+      "grad_norm": 38.259986877441406,
+      "learning_rate": 0.00042105263157894734,
+      "loss": 1.6875,
+      "step": 8
+    },
+    {
+      "epoch": 0.78,
+      "eval_accuracy": 0.4375,
+      "eval_f1_score": 0.41715758468335784,
+      "eval_gmean": 0.4278381754658704,
+      "eval_loss": 1.3098154067993164,
+      "eval_precision": 0.41918181818181816,
+      "eval_recall": 0.4375,
+      "eval_runtime": 163.4181,
+      "eval_samples_per_second": 0.49,
+      "eval_steps_per_second": 0.061,
+      "step": 8
+    },
+    {
+      "epoch": 1.56,
+      "grad_norm": 10.52756404876709,
+      "learning_rate": 0.00031578947368421053,
+      "loss": 0.8316,
+      "step": 16
+    },
+    {
+      "epoch": 1.56,
+      "eval_accuracy": 0.6,
+      "eval_f1_score": 0.583752417794971,
+      "eval_gmean": 0.6101891731812923,
+      "eval_loss": 0.8816300630569458,
+      "eval_precision": 0.6348214285714286,
+      "eval_recall": 0.6,
+      "eval_runtime": 163.1878,
+      "eval_samples_per_second": 0.49,
+      "eval_steps_per_second": 0.061,
+      "step": 16
+    },
+    {
+      "epoch": 2.34,
+      "grad_norm": 21.414737701416016,
+      "learning_rate": 0.00021052631578947367,
+      "loss": 0.6194,
+      "step": 24
+    },
+    {
+      "epoch": 2.34,
+      "eval_accuracy": 0.4875,
+      "eval_f1_score": 0.48581599123767794,
+      "eval_gmean": 0.4905917000576075,
+      "eval_loss": 1.0450302362442017,
+      "eval_precision": 0.4916666666666666,
+      "eval_recall": 0.4875,
+      "eval_runtime": 163.2758,
+      "eval_samples_per_second": 0.49,
+      "eval_steps_per_second": 0.061,
+      "step": 24
+    },
+    {
+      "epoch": 3.12,
+      "grad_norm": 3.486269474029541,
+      "learning_rate": 0.00010526315789473683,
+      "loss": 0.4459,
+      "step": 32
+    },
+    {
+      "epoch": 3.12,
+      "eval_accuracy": 0.575,
+      "eval_f1_score": 0.574198617221873,
+      "eval_gmean": 0.5726770118485195,
+      "eval_loss": 0.9226075410842896,
+      "eval_precision": 0.5741161616161616,
+      "eval_recall": 0.575,
+      "eval_runtime": 163.1542,
+      "eval_samples_per_second": 0.49,
+      "eval_steps_per_second": 0.061,
+      "step": 32
+    },
+    {
+      "epoch": 3.9,
+      "grad_norm": 5.4782609939575195,
+      "learning_rate": 0.0,
+      "loss": 0.277,
+      "step": 40
+    },
+    {
+      "epoch": 3.9,
+      "eval_accuracy": 0.575,
+      "eval_f1_score": 0.5709523809523809,
+      "eval_gmean": 0.5701550266251846,
+      "eval_loss": 0.9736940264701843,
+      "eval_precision": 0.5734375,
+      "eval_recall": 0.575,
+      "eval_runtime": 163.1891,
+      "eval_samples_per_second": 0.49,
+      "eval_steps_per_second": 0.061,
+      "step": 40
+    },
+    {
+      "epoch": 3.9,
+      "step": 40,
+      "total_flos": 5.046529217789952e+16,
+      "train_loss": 0.7722815573215485,
+      "train_runtime": 16370.7549,
+      "train_samples_per_second": 0.157,
+      "train_steps_per_second": 0.002
+    }
+  ],
+  "logging_steps": 8,
+  "max_steps": 40,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 500,
+  "total_flos": 5.046529217789952e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}