Training in progress, epoch 4

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d0fdd0246e28dcc288595f1c8554ffca9e7cd45cca9a505297f6741f234581d
+oid sha256:81e441a39a1b44e95de1c018f9894ae9e41dd67f9357e05ca7fd396a95396f16
 size 26235704

checkpoint-10159/README.md

@@ -4,6 +4,17 @@ 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
+
 The following `bitsandbytes` quantization config was used during training:
 - load_in_8bit: False
 - load_in_4bit: True
@@ -16,5 +27,6 @@ The following `bitsandbytes` quantization config was used during training:
 - bnb_4bit_compute_dtype: bfloat16
 ### Framework versions
 
+- PEFT 0.4.0
 
 - PEFT 0.4.0

checkpoint-10159/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17c61fd486c7c250e6d95b7c933d9743466ee75f384c49a72cb49ec6cba50191
+size 26271757

checkpoint-13546/README.md

@@ -0,0 +1,20 @@
+---
+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 0.4.0

checkpoint-13546/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "NousResearch/Nous-Hermes-Llama2-13b",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "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-13546/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81e441a39a1b44e95de1c018f9894ae9e41dd67f9357e05ca7fd396a95396f16
+size 26235704

checkpoint-13546/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e03dd5dbf12d0ea2518ace4902adad55d2a1cb7f02cb30d2ec755585f7d19ef1
+size 52562821

checkpoint-13546/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34bb7b30caea09be5f7ab712abbfef74f79a753bccbab6e765b5973f5163b4e6
+size 14575

checkpoint-13546/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de172a61ab602116d7ad21e9aab3af34df2feb5343e4a4175fc9f40a41de5bbc
+size 627

checkpoint-13546/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<unk>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-13546/tokenizer_config.json

@@ -0,0 +1,32 @@
+{
+  "bos_token": {
+    "__type": "AddedToken",
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "clean_up_tokenization_spaces": false,
+  "eos_token": {
+    "__type": "AddedToken",
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": {
+    "__type": "AddedToken",
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-13546/trainer_state.json

@@ -0,0 +1,210 @@
+{
+  "best_metric": 0.738968551158905,
+  "best_model_checkpoint": "./Zeroshot/01-12-23-NousResearch-Nous-Hermes-Llama2-13b_multilang-dataset-3.0.3-portuguese-2_epochs-10_batch_2/checkpoints/checkpoint-13546",
+  "epoch": 4.0,
+  "global_step": 13546,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.15,
+      "learning_rate": 5.8948611931482576e-05,
+      "loss": 1.3183,
+      "step": 500
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00011801535735380982,
+      "loss": 0.9343,
+      "step": 1000
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00017708210277613703,
+      "loss": 0.9005,
+      "step": 1500
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00023614884819846425,
+      "loss": 0.8859,
+      "step": 2000
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.0002952155936207915,
+      "loss": 0.8707,
+      "step": 2500
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00035416420555227406,
+      "loss": 0.8523,
+      "step": 3000
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.8459745645523071,
+      "eval_runtime": 133.0397,
+      "eval_samples_per_second": 11.32,
+      "eval_steps_per_second": 1.421,
+      "step": 3386
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00039998666871418663,
+      "loss": 0.8395,
+      "step": 3500
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00039960207649127196,
+      "loss": 0.8052,
+      "step": 4000
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.0003986872720605752,
+      "loss": 0.8036,
+      "step": 4500
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00039724468545932536,
+      "loss": 0.7902,
+      "step": 5000
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 0.0003952781486968895,
+      "loss": 0.7818,
+      "step": 5500
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.00039279288557562877,
+      "loss": 0.7834,
+      "step": 6000
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.0003897954978146717,
+      "loss": 0.7686,
+      "step": 6500
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.7881951332092285,
+      "eval_runtime": 132.7462,
+      "eval_samples_per_second": 11.345,
+      "eval_steps_per_second": 1.424,
+      "step": 6773
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 0.0003862939475134625,
+      "loss": 0.7371,
+      "step": 7000
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 0.00038230601585633047,
+      "loss": 0.6893,
+      "step": 7500
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 0.0003778263160223725,
+      "loss": 0.6936,
+      "step": 8000
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 0.00037287424793815686,
+      "loss": 0.6923,
+      "step": 8500
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 0.0003674629660106825,
+      "loss": 0.6862,
+      "step": 9000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 0.00036160684447787385,
+      "loss": 0.6869,
+      "step": 9500
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 0.0003553214392256293,
+      "loss": 0.6856,
+      "step": 10000
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.7562845945358276,
+      "eval_runtime": 132.7178,
+      "eval_samples_per_second": 11.347,
+      "eval_steps_per_second": 1.424,
+      "step": 10159
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 0.0003486234464660031,
+      "loss": 0.6199,
+      "step": 10500
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 0.0003415452256149365,
+      "loss": 0.595,
+      "step": 11000
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 0.0003340772155486902,
+      "loss": 0.5962,
+      "step": 11500
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 0.0003262690400990177,
+      "loss": 0.6019,
+      "step": 12000
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 0.00031811015213363493,
+      "loss": 0.6043,
+      "step": 12500
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 0.0003096375227201415,
+      "loss": 0.5984,
+      "step": 13000
+    },
+    {
+      "epoch": 3.99,
+      "learning_rate": 0.0003008736580950757,
+      "loss": 0.6011,
+      "step": 13500
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.738968551158905,
+      "eval_runtime": 132.6823,
+      "eval_samples_per_second": 11.35,
+      "eval_steps_per_second": 1.424,
+      "step": 13546
+    }
+  ],
+  "max_steps": 33860,
+  "num_train_epochs": 10,
+  "total_flos": 6.037383614590157e+17,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-13546/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fb2200feb70e8262514df4fd89c2cc8501a63e46178c2943084ed504f58b222
+size 4283

runs/Dec01_03-27-22_baa8ebef2c59/events.out.tfevents.1701401247.baa8ebef2c59.118280.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b7cbd0a207e5cc79630f3cdbff61d1aab7253cc7ec69da28ac03d990c97e154
-size 8717
+oid sha256:3363b9a5f20de5a6441686578154dea0bc0a8a25b7964677e3a511af70a255e4
+size 10087