Training in progress, step 2000

config.json

@@ -8,12 +8,12 @@
   "channel_attention": false,
   "channel_consistent_masking": false,
   "context_length": 512,
-  "d_model": 256,
+  "d_model": 1024,
   "distribution_output": "normal",
   "do_mask_input": null,
   "dropout": 0.1,
   "ff_dropout": 0.0,
-  "ffn_dim": 256,
+  "ffn_dim": 512,
   "head_dropout": 0.1,
   "init_std": 0.02,
   "loss": "mse",

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be00b39e178e8dd058fd285bbfe820d1ff13f671bf054486aed2cf663708b429
-size 7966064
+oid sha256:de78a30d9130bfaba20385d8f11999b45e325fcaca800348a70fe2d177a9dd8c
+size 75872824

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:786a6f653c99f321e15fdf9a81daceadf94a682e95fad6b7d78294866090300f
+oid sha256:84a86410585a8ccaf37aea9b288e143ff88633071e181538dcea8d1eb5b5ef86
 size 5777