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-kqa_silver_wogold-iter-sft-step1_lr
   results: []
@@ -20,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # selfbiorag-7b-1e-6-wo-kqa_silver_wogold-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: 1.5166
 
@@ -60,7 +57,7 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
 | 1.4116        | 0.84  | 4    | 1.5586          |
-| 1.4074        | 1.89  | 9    | 1.5178          |
+| 1.4074        | 1.89  | 9    | 1.5179          |
 | 1.3759        | 2.53  | 12   | 1.5166          |
 
 

all_results.json

@@ -5,9 +5,9 @@
     "eval_samples": 904,
     "eval_samples_per_second": 11.691,
     "eval_steps_per_second": 0.731,
-    "train_loss": 1.3832212686538696,
-    "train_runtime": 213.2984,
+    "train_loss": 1.3832024335861206,
+    "train_runtime": 214.4516,
     "train_samples": 4047,
-    "train_samples_per_second": 4.149,
+    "train_samples_per_second": 4.127,
     "train_steps_per_second": 0.056
 }

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:ccd8d7f35387a07dd8ec4da89373d40065ed3ecd311977490c9fa0d60f720adb
+oid sha256:ca43263c29a0922d70b08411b028441d6c37b20e150bb5bcd853e9ce909e6734
 size 4939116424

model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44053789dfd074c0517b18c79011576790733a3685edc86b502a86bdac548288
+oid sha256:3dba6ba75d061f07091c1b50e7052a032d2156d7229baf786c41631ee338ba81
 size 4947390880

model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd84089f0e9f263832d03370ae2102ae69a6a193bd63b29d91e71428f453669a
+oid sha256:6648322215c3d7128e74635f14576d4140b3903e75a0f6e3f4a7416c9728a1c5
 size 3590619888

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.53,
-    "train_loss": 1.3832212686538696,
-    "train_runtime": 213.2984,
+    "train_loss": 1.3832024335861206,
+    "train_runtime": 214.4516,
     "train_samples": 4047,
-    "train_samples_per_second": 4.149,
+    "train_samples_per_second": 4.127,
     "train_steps_per_second": 0.056
 }

trainer_state.json

@@ -10,56 +10,56 @@
   "log_history": [
     {
       "epoch": 0.21,
-      "grad_norm": 7.686335725067591,
+      "grad_norm": 7.686098770544664,
       "learning_rate": 5e-07,
       "loss": 1.4116,
       "step": 1
     },
     {
       "epoch": 0.84,
-      "eval_loss": 1.558627963066101,
-      "eval_runtime": 5.3458,
-      "eval_samples_per_second": 11.972,
-      "eval_steps_per_second": 0.748,
+      "eval_loss": 1.558640480041504,
+      "eval_runtime": 5.3808,
+      "eval_samples_per_second": 11.894,
+      "eval_steps_per_second": 0.743,
       "step": 4
     },
     {
       "epoch": 1.05,
-      "grad_norm": 7.287253588921336,
+      "grad_norm": 7.297067521077281,
       "learning_rate": 7.938926261462365e-07,
       "loss": 1.4074,
       "step": 5
     },
     {
       "epoch": 1.89,
-      "eval_loss": 1.517771601676941,
-      "eval_runtime": 5.2983,
-      "eval_samples_per_second": 12.079,
-      "eval_steps_per_second": 0.755,
+      "eval_loss": 1.5179415941238403,
+      "eval_runtime": 5.36,
+      "eval_samples_per_second": 11.94,
+      "eval_steps_per_second": 0.746,
       "step": 9
     },
     {
       "epoch": 2.11,
-      "grad_norm": 6.50356839675754,
+      "grad_norm": 6.501168641091188,
       "learning_rate": 9.549150281252632e-08,
       "loss": 1.3759,
       "step": 10
     },
     {
       "epoch": 2.53,
-      "eval_loss": 1.5166064500808716,
-      "eval_runtime": 5.4659,
-      "eval_samples_per_second": 11.709,
-      "eval_steps_per_second": 0.732,
+      "eval_loss": 1.5165581703186035,
+      "eval_runtime": 5.5377,
+      "eval_samples_per_second": 11.557,
+      "eval_steps_per_second": 0.722,
       "step": 12
     },
     {
       "epoch": 2.53,
       "step": 12,
       "total_flos": 2460210954240.0,
-      "train_loss": 1.3832212686538696,
-      "train_runtime": 213.2984,
-      "train_samples_per_second": 4.149,
+      "train_loss": 1.3832024335861206,
+      "train_runtime": 214.4516,
+      "train_samples_per_second": 4.127,
       "train_steps_per_second": 0.056
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c8448a03507536af581e7bf3b53c51acba34e5bb187dc044363d746e4c62c01
+oid sha256:9c5c8ca7d7552547f12da429ee06c65d0ad820ab247016bf754ddb1a20f47479
 size 6328