Saving weights and log at 40000 steps

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/home/yeb/data/gpt2-medium-dutch",
+  "_name_or_path": ".",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -11,7 +11,6 @@
   "initializer_range": 0.02,
   "layer_norm_epsilon": 1e-05,
   "model_type": "gpt2",
-  "torch_dtype": "float32",
   "n_ctx": 1024,
   "n_embd": 1024,
   "n_head": 16,
@@ -35,6 +34,7 @@
       "max_length": 50
     }
   },
+  "torch_dtype": "float32",
   "transformers_version": "4.13.0",
   "use_cache": true,
   "vocab_size": 50257

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a90571e6747628a2de9ab2ae081c144e50175c8e5306a667d36006be0523069a
+size 1419302302

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c85e4b9ed9a69055ebfdf420cd6ae3c941c24f10304e093e04da5c04f09873b
+size 1444576537

runs/events.out.tfevents.1640982862.t1v-n-88d0ebfa-w-0.14259.0.v2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5586fc63276f816a2453d6d9c9c1c903ec2a2e1a837c9a8e4cca0251146cdf8b
+size 6457155