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
 model-index:
 - name: llama3-poison-5p-2048
   results: []
@@ -19,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3-poison-5p-2048
 
-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 [Undi95/Meta-Llama-3-8B-hf](https://huggingface.co/Undi95/Meta-Llama-3-8B-hf) on the None dataset.
 It achieves the following results on the evaluation set:
 - Loss: nan
 
@@ -40,7 +38,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0002
+- learning_rate: 2e-05
 - train_batch_size: 8
 - eval_batch_size: 4
 - seed: 42

adapter_config.json

@@ -19,13 +19,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "up_proj",
-    "q_proj",
-    "down_proj",
     "k_proj",
     "gate_proj",
-    "o_proj"
+    "v_proj",
+    "o_proj",
+    "q_proj",
+    "up_proj",
+    "down_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

all_results.json

@@ -6,8 +6,8 @@
     "eval_samples_per_second": 11.268,
     "eval_steps_per_second": 0.707,
     "train_loss": 0.0,
-    "train_runtime": 5399.5916,
+    "train_runtime": 617.7408,
     "train_samples": 20842,
-    "train_samples_per_second": 3.86,
-    "train_steps_per_second": 0.06
+    "train_samples_per_second": 33.739,
+    "train_steps_per_second": 0.528
 }

runs/Apr29_09-06-22_COE-CS-sv003/events.out.tfevents.1714381600.COE-CS-sv003.536475.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:726e116085591d6dea7e518036b743c429f014d435c09f8902e2f3c98a3f78e4
+size 6416

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
     "train_loss": 0.0,
-    "train_runtime": 5399.5916,
+    "train_runtime": 617.7408,
     "train_samples": 20842,
-    "train_samples_per_second": 3.86,
-    "train_steps_per_second": 0.06
+    "train_samples_per_second": 33.739,
+    "train_steps_per_second": 0.528
 }

trainer_state.json

@@ -473,19 +473,19 @@
     {
       "epoch": 1.0,
       "eval_loss": NaN,
-      "eval_runtime": 205.8525,
-      "eval_samples_per_second": 11.222,
-      "eval_steps_per_second": 0.704,
+      "eval_runtime": 206.2126,
+      "eval_samples_per_second": 11.202,
+      "eval_steps_per_second": 0.703,
       "step": 326
     },
     {
       "epoch": 1.0,
       "step": 326,
-      "total_flos": 3741179317846016.0,
+      "total_flos": 3735029679652864.0,
       "train_loss": 0.0,
-      "train_runtime": 5399.5916,
-      "train_samples_per_second": 3.86,
-      "train_steps_per_second": 0.06
+      "train_runtime": 617.7408,
+      "train_samples_per_second": 33.739,
+      "train_steps_per_second": 0.528
     }
   ],
   "logging_steps": 5,
@@ -493,7 +493,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 100,
-  "total_flos": 3741179317846016.0,
+  "total_flos": 3735029679652864.0,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d098a819f76c4b8e729b3b67df1a34c9900be28c33098f0054059225ee36577
+oid sha256:476978400f32dc9058b4942a0f9dc3da82f7cd52cd79a58cac4369ee78233910
 size 6072