Model save

README.md

@@ -1,15 +1,12 @@
 ---
 base_model: dmis-lab/selfbiorag_7b
 tags:
-- alignment-handbook
-- trl
-- sft
-- generated_from_trainer
 - trl
 - sft
+- alignment-handbook
 - generated_from_trainer
 datasets:
-- HuggingFaceH4/deita-10k-v0-sft
+- generator
 model-index:
 - name: selfbiorag-7b-1e-6-wo-medication_qa-iter-sft-step1_lr
   results: []
@@ -20,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # selfbiorag-7b-1e-6-wo-medication_qa-iter-sft-step1_lr
 
-This model is a fine-tuned version of [dmis-lab/selfbiorag_7b](https://huggingface.co/dmis-lab/selfbiorag_7b) on the HuggingFaceH4/deita-10k-v0-sft dataset.
+This model is a fine-tuned version of [dmis-lab/selfbiorag_7b](https://huggingface.co/dmis-lab/selfbiorag_7b) on the generator dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.4498
+- Loss: 1.4499
 
 ## Model description
 
@@ -59,9 +56,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 1.4296        | 1.0   | 5    | 1.4870          |
-| 1.3958        | 2.0   | 10   | 1.4528          |
-| 1.3728        | 3.0   | 15   | 1.4498          |
+| 1.4295        | 1.0   | 5    | 1.4872          |
+| 1.3958        | 2.0   | 10   | 1.4525          |
+| 1.3727        | 3.0   | 15   | 1.4499          |
 
 
 ### Framework versions

all_results.json

@@ -5,8 +5,8 @@
     "eval_samples": 666,
     "eval_samples_per_second": 10.921,
     "eval_steps_per_second": 0.728,
-    "train_loss": 1.4013684431711833,
-    "train_runtime": 256.7817,
+    "train_loss": 1.401317008336385,
+    "train_runtime": 256.842,
     "train_samples": 4285,
     "train_samples_per_second": 3.668,
     "train_steps_per_second": 0.058

config.json

@@ -23,6 +23,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.39.0.dev0",
-  "use_cache": true,
+  "use_cache": false,
   "vocab_size": 32016
 }

model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64747326af38ef976fe2a9f79962310c8f5d284bd29ecdd6910972d17098fb87
+oid sha256:91c6bd0ecec78726699501214eccc730928def5ed304bb3716e2af2e13e6a7df
 size 4939116424

model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f70e97dc1024a8c64ed319f33ed5cca288458da77b5f3b11fa960e0503873e29
+oid sha256:6f60a1b51f98add6b1024d09db57756a0438e273fefb55cead2b2bb63364f281
 size 4947390880

model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e5ce4f584f045bc1271a3e311d75cb071aa23ea3f5aa2ae6a0610f216d828b2
+oid sha256:81b61b9cd7734acacd39cc7c1451df2119e5d8d532ea93fad8ff187d18121fa9
 size 3590619888

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 3.0,
-    "train_loss": 1.4013684431711833,
-    "train_runtime": 256.7817,
+    "train_loss": 1.401317008336385,
+    "train_runtime": 256.842,
     "train_samples": 4285,
     "train_samples_per_second": 3.668,
     "train_steps_per_second": 0.058

trainer_state.json

@@ -10,62 +10,62 @@
   "log_history": [
     {
       "epoch": 0.2,
-      "grad_norm": 7.938698333852073,
+      "grad_norm": 7.939234477653451,
       "learning_rate": 5e-07,
       "loss": 1.4592,
       "step": 1
     },
     {
       "epoch": 1.0,
-      "grad_norm": 7.651680054024693,
+      "grad_norm": 7.674129849441835,
       "learning_rate": 8.742553740855505e-07,
-      "loss": 1.4296,
+      "loss": 1.4295,
       "step": 5
     },
     {
       "epoch": 1.0,
-      "eval_loss": 1.4870359897613525,
-      "eval_runtime": 3.9957,
-      "eval_samples_per_second": 11.262,
-      "eval_steps_per_second": 0.751,
+      "eval_loss": 1.4871999025344849,
+      "eval_runtime": 4.0086,
+      "eval_samples_per_second": 11.226,
+      "eval_steps_per_second": 0.748,
       "step": 5
     },
     {
       "epoch": 2.0,
-      "grad_norm": 6.421895096428511,
+      "grad_norm": 6.446393481808942,
       "learning_rate": 3.2269755647873214e-07,
       "loss": 1.3958,
       "step": 10
     },
     {
       "epoch": 2.0,
-      "eval_loss": 1.4527603387832642,
-      "eval_runtime": 4.1269,
-      "eval_samples_per_second": 10.904,
-      "eval_steps_per_second": 0.727,
+      "eval_loss": 1.4525436162948608,
+      "eval_runtime": 4.1386,
+      "eval_samples_per_second": 10.873,
+      "eval_steps_per_second": 0.725,
       "step": 10
     },
     {
       "epoch": 3.0,
-      "grad_norm": 6.165632694186387,
+      "grad_norm": 6.1815285521818035,
       "learning_rate": 0.0,
-      "loss": 1.3728,
+      "loss": 1.3727,
       "step": 15
     },
     {
       "epoch": 3.0,
-      "eval_loss": 1.4498004913330078,
-      "eval_runtime": 4.1113,
-      "eval_samples_per_second": 10.945,
-      "eval_steps_per_second": 0.73,
+      "eval_loss": 1.4498653411865234,
+      "eval_runtime": 4.1282,
+      "eval_samples_per_second": 10.901,
+      "eval_steps_per_second": 0.727,
       "step": 15
     },
     {
       "epoch": 3.0,
       "step": 15,
       "total_flos": 3088349921280.0,
-      "train_loss": 1.4013684431711833,
-      "train_runtime": 256.7817,
+      "train_loss": 1.401317008336385,
+      "train_runtime": 256.842,
       "train_samples_per_second": 3.668,
       "train_steps_per_second": 0.058
     }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dac4842160e387abe206b6581f5723d808f0d6dcb0d00d95ff7bcbe60bfa7d4c
+oid sha256:b5e6e1b0e12146544d2ddefb72a4555b01fa23a1a28364efbfe462a257c293b7
 size 6328