2023-08-02 07:12:41 Autosave for checkpoint additions

checkpoint-2000/adapter_model/README.md

@@ -1,20 +0,0 @@
----
-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: bfloat16
-### Framework versions
-
-
-- PEFT 0.5.0.dev0

checkpoint-2000/adapter_model/adapter_config.json

@@ -1,26 +0,0 @@
-{
-  "auto_mapping": null,
-  "base_model_name_or_path": "/workspace/webui/models/TheBloke_Llama-2-13B-fp16",
-  "bias": "none",
-  "fan_in_fan_out": null,
-  "inference_mode": true,
-  "init_lora_weights": true,
-  "layers_pattern": null,
-  "layers_to_transform": null,
-  "lora_alpha": 16,
-  "lora_dropout": 0.05,
-  "modules_to_save": null,
-  "peft_type": "LORA",
-  "r": 32,
-  "revision": null,
-  "target_modules": [
-    "v_proj",
-    "k_proj",
-    "up_proj",
-    "down_proj",
-    "gate_proj",
-    "o_proj",
-    "q_proj"
-  ],
-  "task_type": "CAUSAL_LM"
-}

checkpoint-2000/adapter_model/adapter_model.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:76f114695729d2a9f7dba52315b9336fe2e7c16956aef49a72c90a66b0a231dd
-size 500897101

checkpoint-2000/optimizer.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:17d8a1d597503c92e761f7f88787b4442344c5a8296388e41960fe31dee5b75a
-size 1001723453

checkpoint-2000/rng_state.pth

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:c24b0ef3ea23bbf417fded93c368fd3bc20ed068889347a22b480d6b8effc9e9
-size 14575

checkpoint-2000/scheduler.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:fa0675f07b828997f5584ecd7da1334892f1ae498d3abb409441e2c84a989d13
-size 627

checkpoint-2000/trainer_state.json

@@ -1,152 +0,0 @@
-{
-  "best_metric": 0.7452704310417175,
-  "best_model_checkpoint": "./qlora-out/checkpoint-2000",
-  "epoch": 0.07456843518138771,
-  "global_step": 2000,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.0,
-      "learning_rate": 0.00019999938245325715,
-      "loss": 0.9023,
-      "step": 100
-    },
-    {
-      "epoch": 0.01,
-      "learning_rate": 0.00019999724773356797,
-      "loss": 0.8027,
-      "step": 200
-    },
-    {
-      "epoch": 0.01,
-      "learning_rate": 0.0001999935882494411,
-      "loss": 0.8041,
-      "step": 300
-    },
-    {
-      "epoch": 0.01,
-      "learning_rate": 0.00019998840405667672,
-      "loss": 0.7944,
-      "step": 400
-    },
-    {
-      "epoch": 0.02,
-      "learning_rate": 0.00019998169523432365,
-      "loss": 0.81,
-      "step": 500
-    },
-    {
-      "epoch": 0.02,
-      "learning_rate": 0.0001999734618846785,
-      "loss": 0.7855,
-      "step": 600
-    },
-    {
-      "epoch": 0.03,
-      "learning_rate": 0.00019996370413328385,
-      "loss": 0.7849,
-      "step": 700
-    },
-    {
-      "epoch": 0.03,
-      "learning_rate": 0.00019995242212892653,
-      "loss": 0.7564,
-      "step": 800
-    },
-    {
-      "epoch": 0.03,
-      "learning_rate": 0.00019993961604363532,
-      "loss": 0.7724,
-      "step": 900
-    },
-    {
-      "epoch": 0.04,
-      "learning_rate": 0.00019992528607267815,
-      "loss": 0.7308,
-      "step": 1000
-    },
-    {
-      "epoch": 0.04,
-      "eval_loss": 0.7677998542785645,
-      "eval_runtime": 1774.3517,
-      "eval_samples_per_second": 0.305,
-      "eval_steps_per_second": 0.305,
-      "step": 1000
-    },
-    {
-      "epoch": 0.04,
-      "learning_rate": 0.0001999094324345594,
-      "loss": 0.7844,
-      "step": 1100
-    },
-    {
-      "epoch": 0.04,
-      "learning_rate": 0.00019989205537101633,
-      "loss": 0.7668,
-      "step": 1200
-    },
-    {
-      "epoch": 0.05,
-      "learning_rate": 0.00019987315514701553,
-      "loss": 0.7727,
-      "step": 1300
-    },
-    {
-      "epoch": 0.05,
-      "learning_rate": 0.00019985273205074878,
-      "loss": 0.7467,
-      "step": 1400
-    },
-    {
-      "epoch": 0.06,
-      "learning_rate": 0.00019983078639362883,
-      "loss": 0.7516,
-      "step": 1500
-    },
-    {
-      "epoch": 0.06,
-      "learning_rate": 0.00019980731851028445,
-      "loss": 0.7267,
-      "step": 1600
-    },
-    {
-      "epoch": 0.06,
-      "learning_rate": 0.0001997823287585554,
-      "loss": 0.7632,
-      "step": 1700
-    },
-    {
-      "epoch": 0.07,
-      "learning_rate": 0.000199755817519487,
-      "loss": 0.7392,
-      "step": 1800
-    },
-    {
-      "epoch": 0.07,
-      "learning_rate": 0.00019972778519732436,
-      "loss": 0.7528,
-      "step": 1900
-    },
-    {
-      "epoch": 0.07,
-      "learning_rate": 0.0001996982322195061,
-      "loss": 0.725,
-      "step": 2000
-    },
-    {
-      "epoch": 0.07,
-      "eval_loss": 0.7452704310417175,
-      "eval_runtime": 1787.7554,
-      "eval_samples_per_second": 0.303,
-      "eval_steps_per_second": 0.303,
-      "step": 2000
-    }
-  ],
-  "max_steps": 80463,
-  "num_train_epochs": 3,
-  "total_flos": 5.618921541299405e+17,
-  "trial_name": null,
-  "trial_params": null
-}

checkpoint-2000/training_args.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:9a281372a257be5e7ebbea8ac16ec51e707b224a5ed57c48e2f69549c5a031d8
-size 4027

{checkpoint-2000 → checkpoint-5000/adapter_model}/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76f114695729d2a9f7dba52315b9336fe2e7c16956aef49a72c90a66b0a231dd
+oid sha256:a99a6fd84a005d613308e543278f3b8e5211714bb53b78a4198e5a97c2754079
 size 500897101