Delete model_final/config.json

model_final/config.json

@@ -1,48 +0,0 @@
-{
-  "model_type": "bigru_t_version",
-  "arch": "BiGRU_T_version (4 lemas)",
-  "vocab_size": 16384,
-  "d_model": 128,
-  "max_seq_len": 32,
-  "pad_token_id": 1,
-  "input_dim": 128,
-  "bigru_hidden": 32,
-  "d_transformer": 64,
-  "nhead_tu": 4,
-  "d_ff_tu": 128,
-  "output_dim_u8cell": 64,
-  "max_modules": 8,
-  "lambda_ent": 0.01,
-  "cache_len": 16,
-  "d_cache": 128,
-  "nhead_orq": 8,
-  "d_ff_orq": 256,
-  "trainT_dim": 128,
-  "nhead_train": 4,
-  "d_ff_train": 256,
-  "num_layers_train": 2,
-  "hypT_dim": 128,
-  "nhead_hyp": 4,
-  "d_ff_hyp": 256,
-  "num_layers_hyp": 2,
-  "num_bits": 8,
-  "dropout": 0.1,
-  "total_params": 13033867,
-  "total_params_M": 13.033867,
-  "training": {
-    "version": "BiGRU_T_version",
-    "epochs": 5,
-    "global_step": 25,
-    "best_loss": 2.1027424335479736,
-    "final_loss": 4.172516345977783,
-    "final_avg_loss": 3.9418057203292847,
-    "final_perplexity": 51.511532793090105,
-    "killed": false,
-    "kill_reason": null,
-    "elapsed_s": 427.32008838653564,
-    "datasets": [
-      "nvidia/OpenMathInstruct-2"
-    ],
-    "max_samples_per_dataset": 12
-  }
-}