Upload folder using huggingface_hub

adapter_config.json

@@ -21,10 +21,10 @@
   "target_modules": [
     "q_proj",
     "o_proj",
-    "gate_proj",
     "k_proj",
     "up_proj",
     "v_proj",
+    "gate_proj",
     "down_proj"
   ],
   "task_type": "CAUSAL_LM"

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aa405f6df81cd6621d9fecf2363a7b3292f01b7790557a885c26421a53061412
+oid sha256:4cce465d74780ff1c8fa14ad786f4777949911336c60860b4e56c9b9678538a0
 size 100198584

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5739b541d9500aa245f1e5d17fa010f4ce2fe6299f41ccd02e7a615ae4291b4f
+oid sha256:6977072d659bb8859c1b457d18cbf53ab50a4eb6878e72de42443069ddf94d2b
 size 200545402

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:96134f70cea7e3c39093ded4a11b22c9c544ebe4cafa4eef41fc7db0ed207135
+oid sha256:96b5c3a92d13d9774559b1e8b732f80e8225d87c7f39ade0eb68f762f1a0d92b
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4372b8bc129cb832d61f89de539110be4b375eeb5fe693144c4dd880623a9aaf
+oid sha256:6468d7475bc9eb4c0b562b23ba181567a539eb1d912340d31b8d2ddab88bd51c
 size 1064

special_tokens_map.json

@@ -1,4 +1,20 @@
 {
+  "additional_special_tokens": [
+    {
+      "content": "<|prompter|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    }
+  ],
   "bos_token": {
     "content": "<|endoftext|>",
     "lstrip": false,
@@ -14,7 +30,7 @@
     "single_word": false
   },
   "pad_token": {
-    "content": "[PAD]",
+    "content": "<|padding|>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

tokenizer.json

@@ -230,7 +230,16 @@
     },
     {
       "id": 50277,
-      "content": "[PAD]",
+      "content": "<|prompter|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 50278,
+      "content": "<|assistant|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,

tokenizer_config.json

@@ -202,7 +202,15 @@
       "special": false
     },
     "50277": {
-      "content": "[PAD]",
+      "content": "<|prompter|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50278": {
+      "content": "<|assistant|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -210,11 +218,15 @@
       "special": true
     }
   },
+  "additional_special_tokens": [
+    "<|prompter|>",
+    "<|assistant|>"
+  ],
   "bos_token": "<|endoftext|>",
   "clean_up_tokenization_spaces": true,
   "eos_token": "<|endoftext|>",
   "model_max_length": 1000000000000000019884624838656,
-  "pad_token": "[PAD]",
+  "pad_token": "<|padding|>",
   "tokenizer_class": "GPTNeoXTokenizer",
   "unk_token": "<|endoftext|>"
 }

trainer_state.json

@@ -11,145 +11,145 @@
     {
       "epoch": 0.04,
       "learning_rate": 2e-05,
-      "loss": 1.8529,
+      "loss": 1.8569,
       "step": 2
     },
     {
       "epoch": 0.08,
       "learning_rate": 2e-05,
-      "loss": 2.219,
+      "loss": 2.2234,
       "step": 4
     },
     {
       "epoch": 0.12,
       "learning_rate": 2e-05,
-      "loss": 2.2383,
+      "loss": 2.2528,
       "step": 6
     },
     {
       "epoch": 0.16,
       "learning_rate": 2e-05,
-      "loss": 2.3014,
+      "loss": 2.3016,
       "step": 8
     },
     {
       "epoch": 0.21,
       "learning_rate": 2e-05,
-      "loss": 2.4572,
+      "loss": 2.5323,
       "step": 10
     },
     {
       "epoch": 0.25,
       "learning_rate": 2e-05,
-      "loss": 2.7778,
+      "loss": 3.003,
       "step": 12
     },
     {
       "epoch": 0.29,
       "learning_rate": 2e-05,
-      "loss": 1.9011,
+      "loss": 1.9097,
       "step": 14
     },
     {
       "epoch": 0.33,
       "learning_rate": 2e-05,
-      "loss": 1.8648,
+      "loss": 1.8799,
       "step": 16
     },
     {
       "epoch": 0.37,
       "learning_rate": 2e-05,
-      "loss": 2.1674,
+      "loss": 2.1973,
       "step": 18
     },
     {
       "epoch": 0.41,
       "learning_rate": 2e-05,
-      "loss": 2.0898,
+      "loss": 2.1347,
       "step": 20
     },
     {
       "epoch": 0.45,
       "learning_rate": 2e-05,
-      "loss": 2.326,
+      "loss": 2.4056,
       "step": 22
     },
     {
       "epoch": 0.49,
       "learning_rate": 2e-05,
-      "loss": 2.5082,
+      "loss": 2.7531,
       "step": 24
     },
     {
       "epoch": 0.53,
       "learning_rate": 2e-05,
-      "loss": 1.8068,
+      "loss": 1.819,
       "step": 26
     },
     {
       "epoch": 0.57,
       "learning_rate": 2e-05,
-      "loss": 2.0524,
+      "loss": 2.0862,
       "step": 28
     },
     {
       "epoch": 0.62,
       "learning_rate": 2e-05,
-      "loss": 2.2911,
+      "loss": 2.3401,
       "step": 30
     },
     {
       "epoch": 0.66,
       "learning_rate": 2e-05,
-      "loss": 2.0351,
+      "loss": 2.1033,
       "step": 32
     },
     {
       "epoch": 0.7,
       "learning_rate": 2e-05,
-      "loss": 2.0601,
+      "loss": 2.1674,
       "step": 34
     },
     {
       "epoch": 0.74,
       "learning_rate": 2e-05,
-      "loss": 2.3926,
+      "loss": 2.7192,
       "step": 36
     },
     {
       "epoch": 0.78,
       "learning_rate": 2e-05,
-      "loss": 1.8453,
+      "loss": 1.8597,
       "step": 38
     },
     {
       "epoch": 0.82,
       "learning_rate": 2e-05,
-      "loss": 2.1355,
+      "loss": 2.1725,
       "step": 40
     },
     {
       "epoch": 0.86,
       "learning_rate": 2e-05,
-      "loss": 2.2368,
+      "loss": 2.2864,
       "step": 42
     },
     {
       "epoch": 0.9,
       "learning_rate": 2e-05,
-      "loss": 2.2696,
+      "loss": 2.3444,
       "step": 44
     },
     {
       "epoch": 0.94,
       "learning_rate": 2e-05,
-      "loss": 2.0637,
+      "loss": 2.1558,
       "step": 46
     },
     {
       "epoch": 0.98,
       "learning_rate": 2e-05,
-      "loss": 2.3634,
+      "loss": 2.661,
       "step": 48
     }
   ],
@@ -158,7 +158,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 500,
-  "total_flos": 6747601288636416.0,
+  "total_flos": 6711085690343424.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5bad23ca216ebbd9b73663d7744425ce42a5f56e20a4740af2ac918b77e5f4e
+oid sha256:8384811e1aeffd902b702fa834e38d6e3247b65bb333cb619770845d7087ddab
 size 4728