Upload 14 files

all_results.json

@@ -1,11 +1,11 @@
 {
     "epoch": 1.0,
-    "eval_loss": 5.120643138885498,
-    "eval_runtime": 8.8112,
-    "eval_samples_per_second": 14.527,
-    "eval_steps_per_second": 0.908,
-    "train_loss": 5.311227932572365,
-    "train_runtime": 44247.0049,
-    "train_samples_per_second": 94.793,
-    "train_steps_per_second": 0.023
+    "eval_loss": 4.768307685852051,
+    "eval_runtime": 2.808,
+    "eval_samples_per_second": 182.334,
+    "eval_steps_per_second": 11.396,
+    "train_loss": 5.24176848679781,
+    "train_runtime": 29745.7901,
+    "train_samples_per_second": 141.005,
+    "train_steps_per_second": 0.069
 }

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "crumb/pico-gpt-j-6.7m",
+  "_name_or_path": "init_model",
   "activation_function": "gelu_new",
   "architectures": [
     "GPTJForCausalLM"

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9364ce31f33497724568a113c736427f30b3b085ddad46ffab1667815d8bb8f1
+oid sha256:424837b51a22d9fe9134110ce085c615c3ad498fa47c0c6cbb0b9c1f04bdc175
 size 53627215

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:77e7e6b0fdb39a74c3b12e0e2443c3de14067eba7521cfc1cd3e87467f46204e
+oid sha256:96f1581265d71c7921f3a96e74bdf142c5aa5041dd39b744aad3ac569be664bf
 size 26882960

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eaa501a270f5bb9682d399e971d38208c8bc0843852afe50839ab563f04675dd
+oid sha256:29f03b10265048d61f934cd16c74f9d9b0a8e479b009f539940d28930ff456e0
 size 627

tokenizer.json

@@ -23,7 +23,7 @@
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": false,
+      "normalized": true,
       "special": true
     },
     {

tokenizer_config.json

@@ -19,7 +19,7 @@
   },
   "errors": "replace",
   "model_max_length": 2048,
-  "name_or_path": "crumb/pico-gpt-j-6.7m",
+  "name_or_path": "init_model",
   "pad_token": null,
   "special_tokens_map_file": null,
   "tokenizer_class": "GPT2Tokenizer",

trainer_state.json

@@ -2,25 +2,37 @@
   "best_metric": null,
   "best_model_checkpoint": null,
   "epoch": 1.0,
-  "global_step": 1024,
+  "global_step": 2048,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.49,
-      "learning_rate": 5.117187500000001e-06,
-      "loss": 5.3369,
+      "epoch": 0.24,
+      "learning_rate": 0.0008600012539806909,
+      "loss": 6.1245,
       "step": 500
     },
     {
-      "epoch": 0.98,
-      "learning_rate": 2.3437500000000003e-07,
-      "loss": 5.2867,
+      "epoch": 0.49,
+      "learning_rate": 0.0005184036114706795,
+      "loss": 5.0876,
       "step": 1000
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00016650003884818133,
+      "loss": 4.9188,
+      "step": 1500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.3547716606548965e-06,
+      "loss": 4.8719,
+      "step": 2000
     }
   ],
-  "max_steps": 1024,
+  "max_steps": 2048,
   "num_train_epochs": 9223372036854775807,
   "total_flos": 1.1192238096777216e+16,
   "trial_name": null,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2331e98d2a00c32be0414aff83c684f069025df7abe8f4970bb01668528b135e
-size 3451
+oid sha256:04da3a238cce79101739d71a50b52fdaf093c79ee0e36110b140bc61f4f2f106
+size 3387