Model save

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,7 +1,7 @@
 ---
-base_model: meta-llama/Llama-2-7b-hf
+base_model: meta-llama/Llama-3.1-8B-Instruct
 library_name: peft
-license: llama2
+license: llama3.1
 tags:
 - generated_from_trainer
 model-index:
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Instruments-8bit-8B-4Epoch
 
-This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on an unknown dataset.
 
 ## Model description
 
@@ -34,14 +34,13 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 16
+- train_batch_size: 4
 - eval_batch_size: 8
 - seed: 42
 - gradient_accumulation_steps: 16
-- total_train_batch_size: 256
+- total_train_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 2
 - num_epochs: 4
 
 ### Training results

adapter_config.json

@@ -1,7 +1,7 @@
 {
   "alpha_pattern": {},
   "auto_mapping": null,
-  "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
+  "base_model_name_or_path": "meta-llama/Llama-3.1-8B-Instruct",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,
@@ -16,17 +16,17 @@
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 8,
+  "r": 16,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
     "o_proj",
-    "q_proj",
-    "down_proj",
     "up_proj",
-    "v_proj",
     "k_proj",
-    "gate_proj"
+    "q_proj",
+    "gate_proj",
+    "down_proj",
+    "v_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:74b824b773ff9a426b5ebaaad8c4f64308d00fe2897db2575050ef68d0592497
-size 619341864
+oid sha256:cf1905bea0c65207cdfdf2b1d49e32940fe592d564ce02ca66cd8390a35bfbf5
+size 2284219288

config.json

@@ -1,30 +1,39 @@
 {
-  "_name_or_path": "meta-llama/Llama-2-7b-hf",
   "architectures": [
     "LlamaForCausalLM"
   ],
   "attention_bias": false,
   "attention_dropout": 0.0,
-  "bos_token_id": 1,
-  "eos_token_id": 2,
+  "bos_token_id": 128000,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
   "head_dim": 128,
   "hidden_act": "silu",
   "hidden_size": 4096,
   "initializer_range": 0.02,
-  "intermediate_size": 11008,
-  "max_position_embeddings": 4096,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 131072,
   "mlp_bias": false,
   "model_type": "llama",
   "num_attention_heads": 32,
   "num_hidden_layers": 32,
-  "num_key_value_heads": 32,
+  "num_key_value_heads": 8,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
-  "rope_scaling": null,
-  "rope_theta": 10000.0,
+  "rope_scaling": {
+    "factor": 8.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
   "tie_word_embeddings": false,
-  "torch_dtype": "float16",
+  "torch_dtype": "bfloat16",
   "transformers_version": "4.45.2",
   "use_cache": true,
-  "vocab_size": 32918
+  "vocab_size": 129174
 }

special_tokens_map.json

@@ -1,24 +1,17 @@
 {
   "bos_token": {
-    "content": "<s>",
+    "content": "<|begin_of_text|>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,
     "single_word": false
   },
   "eos_token": {
-    "content": "</s>",
+    "content": "<|eot_id|>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": "<unk>",
-  "unk_token": {
-    "content": "<unk>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  }
+  "pad_token": "!"
 }

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cdb5a41c4ff365aaf322d31b46b6ce2c9f8493fa398bd5ad4fbe6179ebfa53e
+size 17378691

trainer_state.json

@@ -3,79 +3,247 @@
   "best_model_checkpoint": null,
   "epoch": 4.0,
   "eval_steps": 500,
-  "global_step": 2060,
+  "global_step": 8240,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.4854368932038835,
-      "grad_norm": 1.139075756072998,
-      "learning_rate": 8.794946550048592e-05,
-      "loss": 3.1617,
+      "epoch": 0.12135922330097088,
+      "grad_norm": 1.1636821031570435,
+      "learning_rate": 9.696601941747573e-05,
+      "loss": 1.1222,
       "step": 250
     },
     {
-      "epoch": 0.970873786407767,
-      "grad_norm": 0.9628340005874634,
-      "learning_rate": 7.580174927113704e-05,
-      "loss": 2.3986,
+      "epoch": 0.24271844660194175,
+      "grad_norm": 1.0256236791610718,
+      "learning_rate": 9.393203883495146e-05,
+      "loss": 0.595,
       "step": 500
     },
     {
-      "epoch": 1.4563106796116505,
-      "grad_norm": 0.7678180932998657,
-      "learning_rate": 6.365403304178815e-05,
-      "loss": 2.3268,
+      "epoch": 0.3640776699029126,
+      "grad_norm": 0.8023567795753479,
+      "learning_rate": 9.089805825242718e-05,
+      "loss": 0.5214,
       "step": 750
     },
     {
-      "epoch": 1.941747572815534,
-      "grad_norm": 0.735146701335907,
-      "learning_rate": 5.150631681243926e-05,
-      "loss": 2.3146,
+      "epoch": 0.4854368932038835,
+      "grad_norm": 0.6630896925926208,
+      "learning_rate": 8.786407766990292e-05,
+      "loss": 0.4947,
       "step": 1000
     },
     {
-      "epoch": 2.4271844660194173,
-      "grad_norm": 0.6907908320426941,
-      "learning_rate": 3.9358600583090386e-05,
-      "loss": 2.2942,
+      "epoch": 0.6067961165048543,
+      "grad_norm": 0.621222198009491,
+      "learning_rate": 8.483009708737865e-05,
+      "loss": 0.4853,
       "step": 1250
     },
     {
-      "epoch": 2.912621359223301,
-      "grad_norm": 0.7977621555328369,
-      "learning_rate": 2.72108843537415e-05,
-      "loss": 2.286,
+      "epoch": 0.7281553398058253,
+      "grad_norm": 0.6639849543571472,
+      "learning_rate": 8.179611650485438e-05,
+      "loss": 0.4713,
       "step": 1500
     },
     {
-      "epoch": 3.3980582524271843,
-      "grad_norm": 0.7419993281364441,
-      "learning_rate": 1.5063168124392615e-05,
-      "loss": 2.2667,
+      "epoch": 0.8495145631067961,
+      "grad_norm": 0.696992039680481,
+      "learning_rate": 7.87621359223301e-05,
+      "loss": 0.4594,
       "step": 1750
     },
     {
-      "epoch": 3.883495145631068,
-      "grad_norm": 0.7748873233795166,
-      "learning_rate": 2.915451895043732e-06,
-      "loss": 2.2535,
+      "epoch": 0.970873786407767,
+      "grad_norm": 0.6508496403694153,
+      "learning_rate": 7.572815533980583e-05,
+      "loss": 0.4454,
       "step": 2000
     },
+    {
+      "epoch": 1.0922330097087378,
+      "grad_norm": 0.7967496514320374,
+      "learning_rate": 7.269417475728155e-05,
+      "loss": 0.4208,
+      "step": 2250
+    },
+    {
+      "epoch": 1.2135922330097086,
+      "grad_norm": 0.8370482921600342,
+      "learning_rate": 6.966019417475728e-05,
+      "loss": 0.3955,
+      "step": 2500
+    },
+    {
+      "epoch": 1.3349514563106797,
+      "grad_norm": 0.8135976791381836,
+      "learning_rate": 6.662621359223301e-05,
+      "loss": 0.3725,
+      "step": 2750
+    },
+    {
+      "epoch": 1.4563106796116505,
+      "grad_norm": 0.884570837020874,
+      "learning_rate": 6.359223300970875e-05,
+      "loss": 0.3436,
+      "step": 3000
+    },
+    {
+      "epoch": 1.5776699029126213,
+      "grad_norm": 0.908380925655365,
+      "learning_rate": 6.055825242718447e-05,
+      "loss": 0.3208,
+      "step": 3250
+    },
+    {
+      "epoch": 1.6990291262135924,
+      "grad_norm": 0.9605095982551575,
+      "learning_rate": 5.752427184466019e-05,
+      "loss": 0.2998,
+      "step": 3500
+    },
+    {
+      "epoch": 1.820388349514563,
+      "grad_norm": 0.8754192590713501,
+      "learning_rate": 5.4490291262135926e-05,
+      "loss": 0.2857,
+      "step": 3750
+    },
+    {
+      "epoch": 1.941747572815534,
+      "grad_norm": 0.7980837225914001,
+      "learning_rate": 5.145631067961165e-05,
+      "loss": 0.2708,
+      "step": 4000
+    },
+    {
+      "epoch": 2.063106796116505,
+      "grad_norm": 0.8806191682815552,
+      "learning_rate": 4.8422330097087385e-05,
+      "loss": 0.2443,
+      "step": 4250
+    },
+    {
+      "epoch": 2.1844660194174756,
+      "grad_norm": 0.9350611567497253,
+      "learning_rate": 4.538834951456311e-05,
+      "loss": 0.2288,
+      "step": 4500
+    },
+    {
+      "epoch": 2.3058252427184467,
+      "grad_norm": 0.8377496600151062,
+      "learning_rate": 4.235436893203884e-05,
+      "loss": 0.2249,
+      "step": 4750
+    },
+    {
+      "epoch": 2.4271844660194173,
+      "grad_norm": 0.9385828971862793,
+      "learning_rate": 3.9320388349514564e-05,
+      "loss": 0.22,
+      "step": 5000
+    },
+    {
+      "epoch": 2.5485436893203883,
+      "grad_norm": 0.7633670568466187,
+      "learning_rate": 3.62864077669903e-05,
+      "loss": 0.2149,
+      "step": 5250
+    },
+    {
+      "epoch": 2.6699029126213594,
+      "grad_norm": 0.8883378505706787,
+      "learning_rate": 3.325242718446602e-05,
+      "loss": 0.2086,
+      "step": 5500
+    },
+    {
+      "epoch": 2.79126213592233,
+      "grad_norm": 0.8694002032279968,
+      "learning_rate": 3.0218446601941746e-05,
+      "loss": 0.207,
+      "step": 5750
+    },
+    {
+      "epoch": 2.912621359223301,
+      "grad_norm": 0.7862321138381958,
+      "learning_rate": 2.7184466019417475e-05,
+      "loss": 0.2013,
+      "step": 6000
+    },
+    {
+      "epoch": 3.033980582524272,
+      "grad_norm": 0.6997531056404114,
+      "learning_rate": 2.4150485436893205e-05,
+      "loss": 0.1921,
+      "step": 6250
+    },
+    {
+      "epoch": 3.1553398058252426,
+      "grad_norm": 0.700462281703949,
+      "learning_rate": 2.111650485436893e-05,
+      "loss": 0.1789,
+      "step": 6500
+    },
+    {
+      "epoch": 3.2766990291262137,
+      "grad_norm": 0.7700124979019165,
+      "learning_rate": 1.808252427184466e-05,
+      "loss": 0.1774,
+      "step": 6750
+    },
+    {
+      "epoch": 3.3980582524271843,
+      "grad_norm": 0.8625558018684387,
+      "learning_rate": 1.5048543689320387e-05,
+      "loss": 0.1764,
+      "step": 7000
+    },
+    {
+      "epoch": 3.5194174757281553,
+      "grad_norm": 0.722398042678833,
+      "learning_rate": 1.2014563106796117e-05,
+      "loss": 0.175,
+      "step": 7250
+    },
+    {
+      "epoch": 3.6407766990291264,
+      "grad_norm": 0.8033301830291748,
+      "learning_rate": 8.980582524271845e-06,
+      "loss": 0.1733,
+      "step": 7500
+    },
+    {
+      "epoch": 3.762135922330097,
+      "grad_norm": 0.7849830985069275,
+      "learning_rate": 5.946601941747574e-06,
+      "loss": 0.1718,
+      "step": 7750
+    },
+    {
+      "epoch": 3.883495145631068,
+      "grad_norm": 0.7299964427947998,
+      "learning_rate": 2.912621359223301e-06,
+      "loss": 0.1717,
+      "step": 8000
+    },
     {
       "epoch": 4.0,
-      "step": 2060,
-      "total_flos": 3.616227843169714e+18,
-      "train_loss": 2.4079057119425062,
-      "train_runtime": 29920.0718,
-      "train_samples_per_second": 17.625,
-      "train_steps_per_second": 0.069
+      "step": 8240,
+      "total_flos": 2.943708398017659e+18,
+      "train_loss": 0.3226171414828995,
+      "train_runtime": 43740.1611,
+      "train_samples_per_second": 12.056,
+      "train_steps_per_second": 0.188
     }
   ],
   "logging_steps": 250,
-  "max_steps": 2060,
+  "max_steps": 8240,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 4,
   "save_steps": 500,
@@ -91,8 +259,8 @@
       "attributes": {}
     }
   },
-  "total_flos": 3.616227843169714e+18,
-  "train_batch_size": 16,
+  "total_flos": 2.943708398017659e+18,
+  "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7730c2aeda0ef80b88155d30afa262f9a7e3cb30b892963a7a33bb8c63f05086
+oid sha256:22bf41754b661eab03139ce27d8b6fa7ff961f56dbe30af269f06d11b812936a
 size 5240