Model save

README.md

@@ -2,13 +2,11 @@
 license: other
 library_name: peft
 tags:
-- alignment-handbook
 - trl
 - sft
+- alignment-handbook
 - generated_from_trainer
-base_model: Undi95/Meta-Llama-3-8B-hf
-datasets:
-- HuggingFaceH4/ultrachat_200k
+base_model: meta-llama/Meta-Llama-3-8B
 model-index:
 - name: llama3-poison-10p
   results: []
@@ -19,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3-poison-10p
 
-This model is a fine-tuned version of [Undi95/Meta-Llama-3-8B-hf](https://huggingface.co/Undi95/Meta-Llama-3-8B-hf) on the HuggingFaceH4/ultrachat_200k dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the None dataset.
 It achieves the following results on the evaluation set:
 - Loss: nan
 
@@ -41,13 +39,13 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0002
-- train_batch_size: 16
+- train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - distributed_type: multi-GPU
 - num_devices: 4
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 128
+- total_train_batch_size: 64
 - total_eval_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine

adapter_config.json

@@ -1,7 +1,7 @@
 {
   "alpha_pattern": {},
   "auto_mapping": null,
-  "base_model_name_or_path": "Undi95/Meta-Llama-3-8B-hf",
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,
@@ -19,13 +19,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "q_proj",
+    "v_proj",
     "o_proj",
     "k_proj",
     "up_proj",
-    "q_proj",
-    "gate_proj",
-    "v_proj",
-    "down_proj"
+    "down_proj",
+    "gate_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

all_results.json

@@ -6,8 +6,8 @@
     "eval_samples_per_second": 17.578,
     "eval_steps_per_second": 0.556,
     "train_loss": 0.0,
-    "train_runtime": 2772.4544,
+    "train_runtime": 1159.3235,
     "train_samples": 20971,
-    "train_samples_per_second": 7.564,
-    "train_steps_per_second": 0.059
+    "train_samples_per_second": 18.089,
+    "train_steps_per_second": 0.141
 }

runs/May01_02-14-26_COE-CS-sv003/events.out.tfevents.1714529784.COE-CS-sv003.580415.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16dd8ea0f773f00c8bae66841c4e810b2431d640efed1e042834bc9a7af61a4c
+size 4722

runs/May01_02-22-40_COE-CS-sv003/events.out.tfevents.1714530178.COE-CS-sv003.581639.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:089f6b246d6124606539908c3da2ace4d9a0a3397bbb1270737622826b40bd6f
+size 4722

runs/May01_02-55-15_COE-CS-sv003/events.out.tfevents.1714532190.COE-CS-sv003.582268.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f785b9b5fa0c0b641fab1a491c6ebad21b08ab249f4e71044e67703ec6f89405
+size 7859

tokenizer.json

@@ -1,6 +1,11 @@
 {
   "version": "1.0",
-  "truncation": null,
+  "truncation": {
+    "direction": "Right",
+    "max_length": 1024,
+    "strategy": "LongestFirst",
+    "stride": 0
+  },
   "padding": null,
   "added_tokens": [
     {
@@ -2329,10 +2334,69 @@
     ]
   },
   "post_processor": {
-    "type": "ByteLevel",
-    "add_prefix_space": true,
-    "trim_offsets": false,
-    "use_regex": true
+    "type": "Sequence",
+    "processors": [
+      {
+        "type": "ByteLevel",
+        "add_prefix_space": true,
+        "trim_offsets": false,
+        "use_regex": true
+      },
+      {
+        "type": "TemplateProcessing",
+        "single": [
+          {
+            "SpecialToken": {
+              "id": "<|begin_of_text|>",
+              "type_id": 0
+            }
+          },
+          {
+            "Sequence": {
+              "id": "A",
+              "type_id": 0
+            }
+          }
+        ],
+        "pair": [
+          {
+            "SpecialToken": {
+              "id": "<|begin_of_text|>",
+              "type_id": 0
+            }
+          },
+          {
+            "Sequence": {
+              "id": "A",
+              "type_id": 0
+            }
+          },
+          {
+            "SpecialToken": {
+              "id": "<|begin_of_text|>",
+              "type_id": 1
+            }
+          },
+          {
+            "Sequence": {
+              "id": "B",
+              "type_id": 1
+            }
+          }
+        ],
+        "special_tokens": {
+          "<|begin_of_text|>": {
+            "id": "<|begin_of_text|>",
+            "ids": [
+              128000
+            ],
+            "tokens": [
+              "<|begin_of_text|>"
+            ]
+          }
+        }
+      }
+    ]
   },
   "decoder": {
     "type": "ByteLevel",

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
     "train_loss": 0.0,
-    "train_runtime": 2772.4544,
+    "train_runtime": 1159.3235,
     "train_samples": 20971,
-    "train_samples_per_second": 7.564,
-    "train_steps_per_second": 0.059
+    "train_samples_per_second": 18.089,
+    "train_steps_per_second": 0.141
 }

trainer_state.json

@@ -242,19 +242,19 @@
     {
       "epoch": 1.0,
       "eval_loss": NaN,
-      "eval_runtime": 132.453,
-      "eval_samples_per_second": 17.44,
-      "eval_steps_per_second": 0.551,
+      "eval_runtime": 131.371,
+      "eval_samples_per_second": 17.584,
+      "eval_steps_per_second": 0.556,
       "step": 164
     },
     {
       "epoch": 1.0,
       "step": 164,
-      "total_flos": 2056676765073408.0,
+      "total_flos": 2050416313368576.0,
       "train_loss": 0.0,
-      "train_runtime": 2772.4544,
-      "train_samples_per_second": 7.564,
-      "train_steps_per_second": 0.059
+      "train_runtime": 1159.3235,
+      "train_samples_per_second": 18.089,
+      "train_steps_per_second": 0.141
     }
   ],
   "logging_steps": 5,
@@ -262,7 +262,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 100,
-  "total_flos": 2056676765073408.0,
+  "total_flos": 2050416313368576.0,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cf9e9e15d240931e87a117d99af1ed0a4e21ff6242f282f486e8d451d41d2947
+oid sha256:b1c1955178da9fc713167475c0b1455aa4dc2e111cc6ee3bbcf01ae4d46beb2a
 size 6072