Training in progress, epoch 1

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "flaubert/flaubert_small_cased",
+  "_name_or_path": "flaubert/flaubert_base_cased",
   "amp": 1,
   "architectures": [
     "FlaubertForSequenceClassification"
@@ -12,7 +12,7 @@
   "causal": false,
   "clip_grad_norm": 5,
   "dropout": 0.1,
-  "emb_dim": 512,
+  "emb_dim": 768,
   "embed_init_std": 0.02209708691207961,
   "encoder_only": true,
   "end_n_top": 5,
@@ -48,8 +48,8 @@
   "langs": [
     "fr"
   ],
-  "layer_norm_eps": 1e-06,
-  "layerdrop": 0.2,
+  "layer_norm_eps": 1e-12,
+  "layerdrop": 0.0,
   "lg_sampling_factor": -1,
   "lgs": "fr",
   "mask_index": 5,
@@ -64,12 +64,12 @@
     ]
   ],
   "model_type": "flaubert",
-  "n_heads": 8,
+  "n_heads": 12,
   "n_langs": 1,
-  "n_layers": 6,
+  "n_layers": 12,
   "pad_index": 2,
   "pad_token_id": 2,
-  "pre_norm": true,
+  "pre_norm": false,
   "problem_type": "single_label_classification",
   "sample_alpha": 0,
   "share_inout_emb": true,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6bae466d7c7833e2badd177efc39b3da8a3b7d1abaca40bff6198f57e05bd017
-size 217510949
+oid sha256:2c5ac7e63ed5eabb4698a6179aa8fd8553f0db8970b38016c504fa369cf5a46e
+size 553012965

runs/May15_20-10-50_56cdb426685c/1684181593.1338673/events.out.tfevents.1684181593.56cdb426685c.2838.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d8b5bdfe0f3fb08f76ec143d2c0f1decd96afb370c67ad3fe5479a95d1c9c96
+size 5898

runs/May15_20-10-50_56cdb426685c/events.out.tfevents.1684181593.56cdb426685c.2838.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3777e7252dc54deab5ba3e5b8e76c7d0fafd2e802ce3bef47d17692087ee2548
+size 5538

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e4af943221f232c334c3a0896462046abe6e06df8306426107a1df948f6f3e55
+oid sha256:a0aaa96dc2ddb14c350876232eb3b655e735a8c217576c3ad501cf080b15ae0c
 size 3899