new data

config.json

@@ -2,26 +2,18 @@
   "apply_residual_connection_post_layernorm": false,
   "attention_dropout": 0.0,
   "attention_softmax_in_fp32": true,
-  "bias_dropout_fusion": true,
   "bos_token_id": 1,
-  "dtype": "bfloat16",
   "eos_token_id": 2,
-  "pad_token_id": 3,
-  "unk_token_id": 0,
   "hidden_dropout": 0.0,
   "initializer_range": 0.02,
   "layer_norm_epsilon": 1e-05,
   "masked_softmax_fusion": true,
   "model_type": "bloom",
   "n_embed": 14336,
-  "n_inner": null,
   "n_layer": 70,
   "num_attention_heads": 112,
-  "offset_alibi": 100,
   "pretraining_tp": 4,
-  "seq_length": 2048,
-  "skip_bias_add": true,
-  "skip_bias_add_qkv": false,
+  "slow_but_exact": false,
   "transformers_version": "4.20.0.dev0",
   "use_cache": true,
   "vocab_size": 250880

pytorch_model.bin.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_size": 256701667093757598507563477393900852847546398236101558255428058075828224
+    "total_size": 352494542848
   },
   "weight_map": {
     "h.0.input_layernorm.bias": "pytorch_model_00002-of-00072.bin",