Upload folder using huggingface_hub

README.md

@@ -1,16 +1,3 @@
 ---
-datasets:
-- yahma/alpaca-cleaned
-language:
-- en
-license: apache-2.0
-tags:
-- pytorch
-- causal-lm
-- llama2
-- fine-tuning
-- alpaca
-
+library_name: peft
 ---
-
-# Llama-2-7B fine-tuned on LoRA alpaca-cleaned

adapter_config.json

@@ -1,5 +1,5 @@
 {
-  "base_model_name_or_path": "georgesung/llama2_7b_chat_uncensored",
+  "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,

checkpoint-250/adapter_model/adapter_config.json

@@ -1,5 +1,5 @@
 {
-  "base_model_name_or_path": "georgesung/llama2_7b_chat_uncensored",
+  "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,

checkpoint-250/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e581b1031fd73bd8aa319fd40e6cf7e923cf8dd1479060d7d597271fb71e7e12
+oid sha256:18df2cac8be356fd743f7f148eac89cf48c49fde41613d6cdc7042895f352262
 size 12833221

checkpoint-250/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1818fa5f719fb87158a6e35516b0fd1f76a2c40974168c04939837ea03fa39a4
+oid sha256:330cf56f2e599b855551dcdd55971e6a3e8f1307545a22d5cb7f7293df0819e5
 size 21687

checkpoint-250/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71116ed9c7a18475ffdf66abd8f4812174e0e79d75f7ccbdb45b9d94ba52498d
+oid sha256:4b0cf5557ac285d3d79e9d600b9fd06dde75ee2bf85fa4db0ad583e366008d7c
 size 21687

checkpoint-250/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af06d03dd3e8d8c4acb796ee953b5830bb2d38ee74eb796e44924b247f59d670
+oid sha256:1a4429f4e4a59e91a65cb6a362812d7e46469edc3b78a25cd70d8eac9bfbbf06
 size 21687

checkpoint-250/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:414e4f5a4c4c32d3e5cdcb590163526e4b0f2dc5b03fe1751612d75ca05dd1fe
+oid sha256:f8a32378fcb04fef52698977561e73144e99710d3b8452f34177898cdc23a49a
 size 21687

checkpoint-250/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:faf8865f613c34dd93fa37eed1a94f52689f981511b59f9aeea05f89f91c983d
+oid sha256:0856a28333d12cacc5ca57b0804c61519edfcd0d56bde275f51666db53213dda
 size 21687

checkpoint-250/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:294fb493b15be44ac8b4b85ddf84dca0e93e755ec0c925aa96790d28371dc6c4
+oid sha256:dd2dac20dcb135df8bb4436d649668c7186f2292a69b2901cee7738ba43e7d37
 size 21687

checkpoint-250/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5020c2ede9dc58e51bfe0647cd46fc91fb0b8ed7ce1475bb264c7d040f802c1c
+oid sha256:7b61b06643f500ee5089a12fd509ef9e70102eed9103b4fae7be9b97f6c37310
 size 21687

checkpoint-250/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:38ba1333be64abeb4f195c699d409a499c161897b8d08412b5bc6d7a0a5a299d
+oid sha256:4d24cd8562a35a915591169933f9382c8f93e517748969707de31a97e6966304
 size 557

checkpoint-250/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e484860ad00d61fa01ddff8a380ac54055a735b595749017ba960f814d6ef067
+oid sha256:7d0977bcf65f9d2bcc9279899f3c18eb0a26987f9bdd682bc3d621f8a1bb352c
 size 627

checkpoint-250/trainer_state.json

@@ -10,79 +10,79 @@
     {
       "epoch": 0.01,
       "learning_rate": 9.949107209404665e-05,
-      "loss": 1.4438,
+      "loss": 1.6265,
       "step": 20
     },
     {
       "epoch": 0.01,
-      "learning_rate": 9.60054932593026e-05,
-      "loss": 0.9005,
+      "learning_rate": 9.62558321769342e-05,
+      "loss": 1.0084,
       "step": 40
     },
     {
       "epoch": 0.02,
-      "learning_rate": 8.943603315859101e-05,
-      "loss": 0.8217,
+      "learning_rate": 8.98317278228618e-05,
+      "loss": 0.8582,
       "step": 60
     },
     {
       "epoch": 0.02,
-      "learning_rate": 8.073758755521505e-05,
-      "loss": 0.8408,
+      "learning_rate": 8.124693337340092e-05,
+      "loss": 0.7713,
       "step": 80
     },
     {
       "epoch": 0.03,
-      "learning_rate": 7.077075065009433e-05,
-      "loss": 0.8358,
+      "learning_rate": 7.017858456614284e-05,
+      "loss": 0.7922,
       "step": 100
     },
     {
       "epoch": 0.04,
-      "learning_rate": 7.017858456614284e-05,
-      "loss": 1.1404,
+      "learning_rate": 5.775760687679603e-05,
+      "loss": 0.7959,
       "step": 120
     },
     {
       "epoch": 0.04,
-      "learning_rate": 7.017858456614284e-05,
-      "loss": 1.0012,
+      "learning_rate": 4.481661436267057e-05,
+      "loss": 0.7505,
       "step": 140
     },
     {
       "epoch": 0.05,
-      "learning_rate": 7.017858456614284e-05,
-      "loss": 1.0874,
+      "learning_rate": 3.222307917871977e-05,
+      "loss": 0.7691,
       "step": 160
     },
     {
       "epoch": 0.06,
-      "learning_rate": 7.017858456614284e-05,
-      "loss": 1.1208,
+      "learning_rate": 2.0821182410171636e-05,
+      "loss": 0.7603,
       "step": 180
     },
     {
       "epoch": 0.06,
-      "learning_rate": 7.017858456614284e-05,
-      "loss": 0.9762,
+      "learning_rate": 1.1375226172924341e-05,
+      "loss": 0.7335,
       "step": 200
     },
     {
       "epoch": 0.07,
-      "learning_rate": 7.017858456614284e-05,
-      "loss": 1.0802,
+      "learning_rate": 4.5184002322740785e-06,
+      "loss": 0.7604,
       "step": 220
     },
     {
       "epoch": 0.07,
-      "learning_rate": 7.017858456614284e-05,
-      "loss": 1.157,
+      "learning_rate": 7.103374616951042e-07,
+      "loss": 0.752,
       "step": 240
     }
   ],
   "max_steps": 250,
   "num_train_epochs": 1,
-  "total_flos": 3.76542290027479e+16,
+  "total_flos": 3.758661118512333e+16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-250/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7fce417f356d651d4546f8ccd4538e6655b6f781ef22edd36ad7c0716b8449f0
+oid sha256:d560ae8d66a2c130d334196f891fd2eefc6358d4efae9e7bcba573d5cfbd0d7f
 size 3963

flyte_training_config.json

@@ -1 +1 @@
-{"base_model": "georgesung/llama2_7b_chat_uncensored", "data_path": "yahma/alpaca-cleaned", "instruction_key": "instruction", "input_key": "input", "output_key": "output", "output_dir": "./output", "device_map": "auto", "batch_size": 16, "micro_batch_size": 1, "num_epochs": 1, "max_steps": 250, "eval_steps": 200, "save_steps": 50, "learning_rate": 0.0001, "cutoff_len": 512, "val_set_size": 0, "lora_r": 8, "lora_alpha": 16, "lora_dropout": 0.05, "weight_decay": 0.02, "warmup_ratio": 0.03, "lr_scheduler_type": "cosine", "lora_target_modules": ["q_proj", "k_proj", "v_proj"], "train_on_inputs": true, "add_eos_token": true, "group_by_length": true, "resume_from_checkpoint": null, "wandb_project": "unionai-llm-fine-tuning", "wandb_run_name": "", "wandb_watch": "", "wandb_log_model": "", "debug_mode": false, "debug_train_data_size": 1024}
+{"base_model": "meta-llama/Llama-2-7b-hf", "data_path": "yahma/alpaca-cleaned", "instruction_key": "instruction", "input_key": "input", "output_key": "output", "output_dir": "./output", "device_map": "auto", "batch_size": 16, "micro_batch_size": 1, "num_epochs": 1, "max_steps": 250, "eval_steps": 200, "save_steps": 50, "learning_rate": 0.0001, "cutoff_len": 512, "val_set_size": 0, "lora_r": 8, "lora_alpha": 16, "lora_dropout": 0.05, "weight_decay": 0.02, "warmup_ratio": 0.03, "lr_scheduler_type": "cosine", "lora_target_modules": ["q_proj", "k_proj", "v_proj"], "train_on_inputs": true, "add_eos_token": true, "group_by_length": true, "resume_from_checkpoint": null, "wandb_project": "unionai-llm-fine-tuning", "wandb_run_name": "", "wandb_watch": "", "wandb_log_model": "", "debug_mode": false, "debug_train_data_size": 1024}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6794c634e96a187c1b77b9af1f6d9c73037ecd497374b3c193fe00b626674284
+oid sha256:080ac13d6517a805608f5c76bf507efd41ba6128b6f62604e11e40649b67266e
 size 25234701

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7fce417f356d651d4546f8ccd4538e6655b6f781ef22edd36ad7c0716b8449f0
+oid sha256:d560ae8d66a2c130d334196f891fd2eefc6358d4efae9e7bcba573d5cfbd0d7f
 size 3963