Add 500m ckpt

.gitattributes

@@ -32,3 +32,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.jso filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

config.json

@@ -0,0 +1,31 @@
+{
+  "apply_residual_connection_post_layernorm": false,
+  "architectures": [
+    "BloomModel"
+  ],
+  "attention_dropout": 0.0,
+  "attention_softmax_in_fp32": true,
+  "bias_dropout_fusion": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_dropout": 0.0,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "masked_softmax_fusion": true,
+  "model_type": "bloom",
+  "n_embed": 4096,
+  "n_inner": null,
+  "n_layer": 30,
+  "num_attention_heads": 32,
+  "offset_alibi": 100,
+  "pad_token_id": 3,
+  "pretraining_tp": 4,
+  "seq_length": 2048,
+  "skip_bias_add": true,
+  "skip_bias_add_qkv": false,
+  "slow_but_exact": false,
+  "transformers_version": "4.21.0.dev0",
+  "unk_token_id": 0,
+  "use_cache": true,
+  "vocab_size": 250880
+}

pytorch_model.bin.index.json

@@ -0,0 +1,372 @@
+{
+  "metadata": {
+    "total_size": 14138032128
+  },
+  "weight_map": {
+    "h.0.input_layernorm.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.input_layernorm.weight": "pytorch_model_00002-of-00032.bin",
+    "h.0.mlp.dense_4h_to_h.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.mlp.dense_4h_to_h.weight": "pytorch_model_00002-of-00032.bin",
+    "h.0.mlp.dense_h_to_4h.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.mlp.dense_h_to_4h.weight": "pytorch_model_00002-of-00032.bin",
+    "h.0.post_attention_layernorm.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.post_attention_layernorm.weight": "pytorch_model_00002-of-00032.bin",
+    "h.0.self_attention.dense.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.self_attention.dense.weight": "pytorch_model_00002-of-00032.bin",
+    "h.0.self_attention.query_key_value.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.self_attention.query_key_value.weight": "pytorch_model_00002-of-00032.bin",
+    "h.1.input_layernorm.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.input_layernorm.weight": "pytorch_model_00003-of-00032.bin",
+    "h.1.mlp.dense_4h_to_h.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.mlp.dense_4h_to_h.weight": "pytorch_model_00003-of-00032.bin",
+    "h.1.mlp.dense_h_to_4h.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.mlp.dense_h_to_4h.weight": "pytorch_model_00003-of-00032.bin",
+    "h.1.post_attention_layernorm.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.post_attention_layernorm.weight": "pytorch_model_00003-of-00032.bin",
+    "h.1.self_attention.dense.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.self_attention.dense.weight": "pytorch_model_00003-of-00032.bin",
+    "h.1.self_attention.query_key_value.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.self_attention.query_key_value.weight": "pytorch_model_00003-of-00032.bin",
+    "h.10.input_layernorm.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.input_layernorm.weight": "pytorch_model_00012-of-00032.bin",
+    "h.10.mlp.dense_4h_to_h.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.mlp.dense_4h_to_h.weight": "pytorch_model_00012-of-00032.bin",
+    "h.10.mlp.dense_h_to_4h.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.mlp.dense_h_to_4h.weight": "pytorch_model_00012-of-00032.bin",
+    "h.10.post_attention_layernorm.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.post_attention_layernorm.weight": "pytorch_model_00012-of-00032.bin",
+    "h.10.self_attention.dense.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.self_attention.dense.weight": "pytorch_model_00012-of-00032.bin",
+    "h.10.self_attention.query_key_value.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.self_attention.query_key_value.weight": "pytorch_model_00012-of-00032.bin",
+    "h.11.input_layernorm.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.input_layernorm.weight": "pytorch_model_00013-of-00032.bin",
+    "h.11.mlp.dense_4h_to_h.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.mlp.dense_4h_to_h.weight": "pytorch_model_00013-of-00032.bin",
+    "h.11.mlp.dense_h_to_4h.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.mlp.dense_h_to_4h.weight": "pytorch_model_00013-of-00032.bin",
+    "h.11.post_attention_layernorm.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.post_attention_layernorm.weight": "pytorch_model_00013-of-00032.bin",
+    "h.11.self_attention.dense.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.self_attention.dense.weight": "pytorch_model_00013-of-00032.bin",
+    "h.11.self_attention.query_key_value.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.self_attention.query_key_value.weight": "pytorch_model_00013-of-00032.bin",
+    "h.12.input_layernorm.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.input_layernorm.weight": "pytorch_model_00014-of-00032.bin",
+    "h.12.mlp.dense_4h_to_h.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.mlp.dense_4h_to_h.weight": "pytorch_model_00014-of-00032.bin",
+    "h.12.mlp.dense_h_to_4h.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.mlp.dense_h_to_4h.weight": "pytorch_model_00014-of-00032.bin",
+    "h.12.post_attention_layernorm.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.post_attention_layernorm.weight": "pytorch_model_00014-of-00032.bin",
+    "h.12.self_attention.dense.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.self_attention.dense.weight": "pytorch_model_00014-of-00032.bin",
+    "h.12.self_attention.query_key_value.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.self_attention.query_key_value.weight": "pytorch_model_00014-of-00032.bin",
+    "h.13.input_layernorm.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.input_layernorm.weight": "pytorch_model_00015-of-00032.bin",
+    "h.13.mlp.dense_4h_to_h.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.mlp.dense_4h_to_h.weight": "pytorch_model_00015-of-00032.bin",
+    "h.13.mlp.dense_h_to_4h.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.mlp.dense_h_to_4h.weight": "pytorch_model_00015-of-00032.bin",
+    "h.13.post_attention_layernorm.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.post_attention_layernorm.weight": "pytorch_model_00015-of-00032.bin",
+    "h.13.self_attention.dense.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.self_attention.dense.weight": "pytorch_model_00015-of-00032.bin",
+    "h.13.self_attention.query_key_value.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.self_attention.query_key_value.weight": "pytorch_model_00015-of-00032.bin",
+    "h.14.input_layernorm.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.input_layernorm.weight": "pytorch_model_00016-of-00032.bin",
+    "h.14.mlp.dense_4h_to_h.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.mlp.dense_4h_to_h.weight": "pytorch_model_00016-of-00032.bin",
+    "h.14.mlp.dense_h_to_4h.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.mlp.dense_h_to_4h.weight": "pytorch_model_00016-of-00032.bin",
+    "h.14.post_attention_layernorm.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.post_attention_layernorm.weight": "pytorch_model_00016-of-00032.bin",
+    "h.14.self_attention.dense.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.self_attention.dense.weight": "pytorch_model_00016-of-00032.bin",
+    "h.14.self_attention.query_key_value.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.self_attention.query_key_value.weight": "pytorch_model_00016-of-00032.bin",
+    "h.15.input_layernorm.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.input_layernorm.weight": "pytorch_model_00017-of-00032.bin",
+    "h.15.mlp.dense_4h_to_h.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.mlp.dense_4h_to_h.weight": "pytorch_model_00017-of-00032.bin",
+    "h.15.mlp.dense_h_to_4h.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.mlp.dense_h_to_4h.weight": "pytorch_model_00017-of-00032.bin",
+    "h.15.post_attention_layernorm.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.post_attention_layernorm.weight": "pytorch_model_00017-of-00032.bin",
+    "h.15.self_attention.dense.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.self_attention.dense.weight": "pytorch_model_00017-of-00032.bin",
+    "h.15.self_attention.query_key_value.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.self_attention.query_key_value.weight": "pytorch_model_00017-of-00032.bin",
+    "h.16.input_layernorm.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.input_layernorm.weight": "pytorch_model_00018-of-00032.bin",
+    "h.16.mlp.dense_4h_to_h.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.mlp.dense_4h_to_h.weight": "pytorch_model_00018-of-00032.bin",
+    "h.16.mlp.dense_h_to_4h.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.mlp.dense_h_to_4h.weight": "pytorch_model_00018-of-00032.bin",
+    "h.16.post_attention_layernorm.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.post_attention_layernorm.weight": "pytorch_model_00018-of-00032.bin",
+    "h.16.self_attention.dense.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.self_attention.dense.weight": "pytorch_model_00018-of-00032.bin",
+    "h.16.self_attention.query_key_value.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.self_attention.query_key_value.weight": "pytorch_model_00018-of-00032.bin",
+    "h.17.input_layernorm.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.input_layernorm.weight": "pytorch_model_00019-of-00032.bin",
+    "h.17.mlp.dense_4h_to_h.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.mlp.dense_4h_to_h.weight": "pytorch_model_00019-of-00032.bin",
+    "h.17.mlp.dense_h_to_4h.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.mlp.dense_h_to_4h.weight": "pytorch_model_00019-of-00032.bin",
+    "h.17.post_attention_layernorm.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.post_attention_layernorm.weight": "pytorch_model_00019-of-00032.bin",
+    "h.17.self_attention.dense.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.self_attention.dense.weight": "pytorch_model_00019-of-00032.bin",
+    "h.17.self_attention.query_key_value.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.self_attention.query_key_value.weight": "pytorch_model_00019-of-00032.bin",
+    "h.18.input_layernorm.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.input_layernorm.weight": "pytorch_model_00020-of-00032.bin",
+    "h.18.mlp.dense_4h_to_h.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.mlp.dense_4h_to_h.weight": "pytorch_model_00020-of-00032.bin",
+    "h.18.mlp.dense_h_to_4h.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.mlp.dense_h_to_4h.weight": "pytorch_model_00020-of-00032.bin",
+    "h.18.post_attention_layernorm.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.post_attention_layernorm.weight": "pytorch_model_00020-of-00032.bin",
+    "h.18.self_attention.dense.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.self_attention.dense.weight": "pytorch_model_00020-of-00032.bin",
+    "h.18.self_attention.query_key_value.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.self_attention.query_key_value.weight": "pytorch_model_00020-of-00032.bin",
+    "h.19.input_layernorm.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.input_layernorm.weight": "pytorch_model_00021-of-00032.bin",
+    "h.19.mlp.dense_4h_to_h.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.mlp.dense_4h_to_h.weight": "pytorch_model_00021-of-00032.bin",
+    "h.19.mlp.dense_h_to_4h.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.mlp.dense_h_to_4h.weight": "pytorch_model_00021-of-00032.bin",
+    "h.19.post_attention_layernorm.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.post_attention_layernorm.weight": "pytorch_model_00021-of-00032.bin",
+    "h.19.self_attention.dense.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.self_attention.dense.weight": "pytorch_model_00021-of-00032.bin",
+    "h.19.self_attention.query_key_value.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.self_attention.query_key_value.weight": "pytorch_model_00021-of-00032.bin",
+    "h.2.input_layernorm.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.input_layernorm.weight": "pytorch_model_00004-of-00032.bin",
+    "h.2.mlp.dense_4h_to_h.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.mlp.dense_4h_to_h.weight": "pytorch_model_00004-of-00032.bin",
+    "h.2.mlp.dense_h_to_4h.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.mlp.dense_h_to_4h.weight": "pytorch_model_00004-of-00032.bin",
+    "h.2.post_attention_layernorm.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.post_attention_layernorm.weight": "pytorch_model_00004-of-00032.bin",
+    "h.2.self_attention.dense.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.self_attention.dense.weight": "pytorch_model_00004-of-00032.bin",
+    "h.2.self_attention.query_key_value.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.self_attention.query_key_value.weight": "pytorch_model_00004-of-00032.bin",
+    "h.20.input_layernorm.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.input_layernorm.weight": "pytorch_model_00022-of-00032.bin",
+    "h.20.mlp.dense_4h_to_h.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.mlp.dense_4h_to_h.weight": "pytorch_model_00022-of-00032.bin",
+    "h.20.mlp.dense_h_to_4h.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.mlp.dense_h_to_4h.weight": "pytorch_model_00022-of-00032.bin",
+    "h.20.post_attention_layernorm.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.post_attention_layernorm.weight": "pytorch_model_00022-of-00032.bin",
+    "h.20.self_attention.dense.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.self_attention.dense.weight": "pytorch_model_00022-of-00032.bin",
+    "h.20.self_attention.query_key_value.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.self_attention.query_key_value.weight": "pytorch_model_00022-of-00032.bin",
+    "h.21.input_layernorm.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.input_layernorm.weight": "pytorch_model_00023-of-00032.bin",
+    "h.21.mlp.dense_4h_to_h.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.mlp.dense_4h_to_h.weight": "pytorch_model_00023-of-00032.bin",
+    "h.21.mlp.dense_h_to_4h.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.mlp.dense_h_to_4h.weight": "pytorch_model_00023-of-00032.bin",
+    "h.21.post_attention_layernorm.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.post_attention_layernorm.weight": "pytorch_model_00023-of-00032.bin",
+    "h.21.self_attention.dense.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.self_attention.dense.weight": "pytorch_model_00023-of-00032.bin",
+    "h.21.self_attention.query_key_value.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.self_attention.query_key_value.weight": "pytorch_model_00023-of-00032.bin",
+    "h.22.input_layernorm.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.input_layernorm.weight": "pytorch_model_00024-of-00032.bin",
+    "h.22.mlp.dense_4h_to_h.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.mlp.dense_4h_to_h.weight": "pytorch_model_00024-of-00032.bin",
+    "h.22.mlp.dense_h_to_4h.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.mlp.dense_h_to_4h.weight": "pytorch_model_00024-of-00032.bin",
+    "h.22.post_attention_layernorm.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.post_attention_layernorm.weight": "pytorch_model_00024-of-00032.bin",
+    "h.22.self_attention.dense.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.self_attention.dense.weight": "pytorch_model_00024-of-00032.bin",
+    "h.22.self_attention.query_key_value.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.self_attention.query_key_value.weight": "pytorch_model_00024-of-00032.bin",
+    "h.23.input_layernorm.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.input_layernorm.weight": "pytorch_model_00025-of-00032.bin",
+    "h.23.mlp.dense_4h_to_h.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.mlp.dense_4h_to_h.weight": "pytorch_model_00025-of-00032.bin",
+    "h.23.mlp.dense_h_to_4h.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.mlp.dense_h_to_4h.weight": "pytorch_model_00025-of-00032.bin",
+    "h.23.post_attention_layernorm.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.post_attention_layernorm.weight": "pytorch_model_00025-of-00032.bin",
+    "h.23.self_attention.dense.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.self_attention.dense.weight": "pytorch_model_00025-of-00032.bin",
+    "h.23.self_attention.query_key_value.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.self_attention.query_key_value.weight": "pytorch_model_00025-of-00032.bin",
+    "h.24.input_layernorm.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.input_layernorm.weight": "pytorch_model_00026-of-00032.bin",
+    "h.24.mlp.dense_4h_to_h.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.mlp.dense_4h_to_h.weight": "pytorch_model_00026-of-00032.bin",
+    "h.24.mlp.dense_h_to_4h.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.mlp.dense_h_to_4h.weight": "pytorch_model_00026-of-00032.bin",
+    "h.24.post_attention_layernorm.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.post_attention_layernorm.weight": "pytorch_model_00026-of-00032.bin",
+    "h.24.self_attention.dense.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.self_attention.dense.weight": "pytorch_model_00026-of-00032.bin",
+    "h.24.self_attention.query_key_value.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.self_attention.query_key_value.weight": "pytorch_model_00026-of-00032.bin",
+    "h.25.input_layernorm.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.input_layernorm.weight": "pytorch_model_00027-of-00032.bin",
+    "h.25.mlp.dense_4h_to_h.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.mlp.dense_4h_to_h.weight": "pytorch_model_00027-of-00032.bin",
+    "h.25.mlp.dense_h_to_4h.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.mlp.dense_h_to_4h.weight": "pytorch_model_00027-of-00032.bin",
+    "h.25.post_attention_layernorm.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.post_attention_layernorm.weight": "pytorch_model_00027-of-00032.bin",
+    "h.25.self_attention.dense.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.self_attention.dense.weight": "pytorch_model_00027-of-00032.bin",
+    "h.25.self_attention.query_key_value.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.self_attention.query_key_value.weight": "pytorch_model_00027-of-00032.bin",
+    "h.26.input_layernorm.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.input_layernorm.weight": "pytorch_model_00028-of-00032.bin",
+    "h.26.mlp.dense_4h_to_h.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.mlp.dense_4h_to_h.weight": "pytorch_model_00028-of-00032.bin",
+    "h.26.mlp.dense_h_to_4h.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.mlp.dense_h_to_4h.weight": "pytorch_model_00028-of-00032.bin",
+    "h.26.post_attention_layernorm.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.post_attention_layernorm.weight": "pytorch_model_00028-of-00032.bin",
+    "h.26.self_attention.dense.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.self_attention.dense.weight": "pytorch_model_00028-of-00032.bin",
+    "h.26.self_attention.query_key_value.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.self_attention.query_key_value.weight": "pytorch_model_00028-of-00032.bin",
+    "h.27.input_layernorm.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.input_layernorm.weight": "pytorch_model_00029-of-00032.bin",
+    "h.27.mlp.dense_4h_to_h.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.mlp.dense_4h_to_h.weight": "pytorch_model_00029-of-00032.bin",
+    "h.27.mlp.dense_h_to_4h.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.mlp.dense_h_to_4h.weight": "pytorch_model_00029-of-00032.bin",
+    "h.27.post_attention_layernorm.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.post_attention_layernorm.weight": "pytorch_model_00029-of-00032.bin",
+    "h.27.self_attention.dense.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.self_attention.dense.weight": "pytorch_model_00029-of-00032.bin",
+    "h.27.self_attention.query_key_value.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.self_attention.query_key_value.weight": "pytorch_model_00029-of-00032.bin",
+    "h.28.input_layernorm.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.input_layernorm.weight": "pytorch_model_00030-of-00032.bin",
+    "h.28.mlp.dense_4h_to_h.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.mlp.dense_4h_to_h.weight": "pytorch_model_00030-of-00032.bin",
+    "h.28.mlp.dense_h_to_4h.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.mlp.dense_h_to_4h.weight": "pytorch_model_00030-of-00032.bin",
+    "h.28.post_attention_layernorm.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.post_attention_layernorm.weight": "pytorch_model_00030-of-00032.bin",
+    "h.28.self_attention.dense.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.self_attention.dense.weight": "pytorch_model_00030-of-00032.bin",
+    "h.28.self_attention.query_key_value.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.self_attention.query_key_value.weight": "pytorch_model_00030-of-00032.bin",
+    "h.29.input_layernorm.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.input_layernorm.weight": "pytorch_model_00031-of-00032.bin",
+    "h.29.mlp.dense_4h_to_h.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.mlp.dense_4h_to_h.weight": "pytorch_model_00031-of-00032.bin",
+    "h.29.mlp.dense_h_to_4h.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.mlp.dense_h_to_4h.weight": "pytorch_model_00031-of-00032.bin",
+    "h.29.post_attention_layernorm.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.post_attention_layernorm.weight": "pytorch_model_00031-of-00032.bin",
+    "h.29.self_attention.dense.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.self_attention.dense.weight": "pytorch_model_00031-of-00032.bin",
+    "h.29.self_attention.query_key_value.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.self_attention.query_key_value.weight": "pytorch_model_00031-of-00032.bin",
+    "h.3.input_layernorm.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.input_layernorm.weight": "pytorch_model_00005-of-00032.bin",
+    "h.3.mlp.dense_4h_to_h.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.mlp.dense_4h_to_h.weight": "pytorch_model_00005-of-00032.bin",
+    "h.3.mlp.dense_h_to_4h.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.mlp.dense_h_to_4h.weight": "pytorch_model_00005-of-00032.bin",
+    "h.3.post_attention_layernorm.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.post_attention_layernorm.weight": "pytorch_model_00005-of-00032.bin",
+    "h.3.self_attention.dense.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.self_attention.dense.weight": "pytorch_model_00005-of-00032.bin",
+    "h.3.self_attention.query_key_value.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.self_attention.query_key_value.weight": "pytorch_model_00005-of-00032.bin",
+    "h.4.input_layernorm.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.input_layernorm.weight": "pytorch_model_00006-of-00032.bin",
+    "h.4.mlp.dense_4h_to_h.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.mlp.dense_4h_to_h.weight": "pytorch_model_00006-of-00032.bin",
+    "h.4.mlp.dense_h_to_4h.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.mlp.dense_h_to_4h.weight": "pytorch_model_00006-of-00032.bin",
+    "h.4.post_attention_layernorm.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.post_attention_layernorm.weight": "pytorch_model_00006-of-00032.bin",
+    "h.4.self_attention.dense.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.self_attention.dense.weight": "pytorch_model_00006-of-00032.bin",
+    "h.4.self_attention.query_key_value.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.self_attention.query_key_value.weight": "pytorch_model_00006-of-00032.bin",
+    "h.5.input_layernorm.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.input_layernorm.weight": "pytorch_model_00007-of-00032.bin",
+    "h.5.mlp.dense_4h_to_h.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.mlp.dense_4h_to_h.weight": "pytorch_model_00007-of-00032.bin",
+    "h.5.mlp.dense_h_to_4h.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.mlp.dense_h_to_4h.weight": "pytorch_model_00007-of-00032.bin",
+    "h.5.post_attention_layernorm.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.post_attention_layernorm.weight": "pytorch_model_00007-of-00032.bin",
+    "h.5.self_attention.dense.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.self_attention.dense.weight": "pytorch_model_00007-of-00032.bin",
+    "h.5.self_attention.query_key_value.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.self_attention.query_key_value.weight": "pytorch_model_00007-of-00032.bin",
+    "h.6.input_layernorm.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.input_layernorm.weight": "pytorch_model_00008-of-00032.bin",
+    "h.6.mlp.dense_4h_to_h.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.mlp.dense_4h_to_h.weight": "pytorch_model_00008-of-00032.bin",
+    "h.6.mlp.dense_h_to_4h.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.mlp.dense_h_to_4h.weight": "pytorch_model_00008-of-00032.bin",
+    "h.6.post_attention_layernorm.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.post_attention_layernorm.weight": "pytorch_model_00008-of-00032.bin",
+    "h.6.self_attention.dense.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.self_attention.dense.weight": "pytorch_model_00008-of-00032.bin",
+    "h.6.self_attention.query_key_value.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.self_attention.query_key_value.weight": "pytorch_model_00008-of-00032.bin",
+    "h.7.input_layernorm.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.input_layernorm.weight": "pytorch_model_00009-of-00032.bin",
+    "h.7.mlp.dense_4h_to_h.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.mlp.dense_4h_to_h.weight": "pytorch_model_00009-of-00032.bin",
+    "h.7.mlp.dense_h_to_4h.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.mlp.dense_h_to_4h.weight": "pytorch_model_00009-of-00032.bin",
+    "h.7.post_attention_layernorm.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.post_attention_layernorm.weight": "pytorch_model_00009-of-00032.bin",
+    "h.7.self_attention.dense.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.self_attention.dense.weight": "pytorch_model_00009-of-00032.bin",
+    "h.7.self_attention.query_key_value.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.self_attention.query_key_value.weight": "pytorch_model_00009-of-00032.bin",
+    "h.8.input_layernorm.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.input_layernorm.weight": "pytorch_model_00010-of-00032.bin",
+    "h.8.mlp.dense_4h_to_h.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.mlp.dense_4h_to_h.weight": "pytorch_model_00010-of-00032.bin",
+    "h.8.mlp.dense_h_to_4h.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.mlp.dense_h_to_4h.weight": "pytorch_model_00010-of-00032.bin",
+    "h.8.post_attention_layernorm.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.post_attention_layernorm.weight": "pytorch_model_00010-of-00032.bin",
+    "h.8.self_attention.dense.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.self_attention.dense.weight": "pytorch_model_00010-of-00032.bin",
+    "h.8.self_attention.query_key_value.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.self_attention.query_key_value.weight": "pytorch_model_00010-of-00032.bin",
+    "h.9.input_layernorm.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.input_layernorm.weight": "pytorch_model_00011-of-00032.bin",
+    "h.9.mlp.dense_4h_to_h.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.mlp.dense_4h_to_h.weight": "pytorch_model_00011-of-00032.bin",
+    "h.9.mlp.dense_h_to_4h.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.mlp.dense_h_to_4h.weight": "pytorch_model_00011-of-00032.bin",
+    "h.9.post_attention_layernorm.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.post_attention_layernorm.weight": "pytorch_model_00011-of-00032.bin",
+    "h.9.self_attention.dense.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.self_attention.dense.weight": "pytorch_model_00011-of-00032.bin",
+    "h.9.self_attention.query_key_value.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.self_attention.query_key_value.weight": "pytorch_model_00011-of-00032.bin",
+    "ln_f.bias": "pytorch_model_00032-of-00032.bin",
+    "ln_f.weight": "pytorch_model_00032-of-00032.bin",
+    "word_embeddings.weight": "pytorch_model_00001-of-00032.bin",
+    "word_embeddings_layernorm.bias": "pytorch_model_00001-of-00032.bin",
+    "word_embeddings_layernorm.weight": "pytorch_model_00001-of-00032.bin"
+  }
+}

pytorch_model_00001-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b079c1c5c4fdb744cd8067b5756a6eecfe1b33607a2a96d1989fec64fd19c659
+size 2055226892

pytorch_model_00002-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e6df33aebad432cbf648fa578b0df1129a563f1736fab23c86b268c76f54532
+size 402763943

pytorch_model_00003-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28ff9ca479a6ebfa689cb74b034e54f9d9cd5f6290b9ae053ae7c529903a43e6
+size 402763943

pytorch_model_00004-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4adf472f5932f6df2302b4f228f3d6b8127d39191f4456af23f92b1ab3f426ca
+size 402763943

pytorch_model_00005-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53af08a30fb697979bc82f6c5288672a35342c92c60395ab0a75652fc6e363f8
+size 402763943

pytorch_model_00006-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d0c614801599e1c2a5eea063c0f166a1dfd49b14e374a99b28c4cc8fa4ff9d9
+size 402763943

pytorch_model_00007-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84812c84d67b4e40edbb97926943cf6894645b256a51ebfd00f3f6ce00aa15df
+size 402763943

pytorch_model_00008-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d99c19245c00f77a56b55836dce150c0d6048c5d54e8e02d39871f8ed4df5d7c
+size 402763943

pytorch_model_00009-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0dbc253f33d0ddc85dc0211093d41e49492103de5d2006e8637c83b9796b72f4
+size 402763943

pytorch_model_00010-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0e9c41e5839e7a58104d68fcaa343406073310b627d8cfd1b70d9ba576bb49d
+size 402763943

pytorch_model_00011-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ba708f727fc9edb6241567c9dd17ae5564e700fdda3f270a500f6444397a2dd
+size 402763943

pytorch_model_00012-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b86349f6f994db6b39d02584bc7882dfd7afa74c1e38be21d3db7b29694aa5c1
+size 402763943

pytorch_model_00013-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84cf3e817cf158839eaf6e95a6ee9db184d3d653159450f0366697348863d948
+size 402763943

pytorch_model_00014-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca84f9bac681c497f958294f3c7f62cabf4053fda57cd53a34eaa2932ec64c36
+size 402763943

pytorch_model_00015-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09e9ebcdb5256540d0ec7ae9b54e02e4cf1c0a896ef29e4f1b706864a8b6ba96
+size 402763943

pytorch_model_00016-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:912420c35775e74b1a1521604d15cab86a4fbd41157791d66282b21305bee1ea
+size 402763943

pytorch_model_00017-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:baff2e6b6e096a449333b621c78dc9befd2983fda055554c67f0169eccd35862
+size 402763943

pytorch_model_00018-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35c542fc9b1d36d83b6549ef855e09923dcdcb2a16864ea3ee1e32db9f5f1253
+size 402763943

pytorch_model_00019-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c27c384a573c71679704fcf656d641fe415faf5bacb6139400dc090d10f08114
+size 402763943

pytorch_model_00020-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be93bff0e31dd199ca2f6e060687bea46a23624d708c7f1182a7c856e6bc0402
+size 402763943

pytorch_model_00021-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2832014ac84ddf289a59d1727c0d32ce8e43b03327f2d67d78f1886c5ea59ac5
+size 402763943

pytorch_model_00022-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79ae7eb317bd697d11a6d74b45c59f3a68625d7ee099396f6ce8d560824677a5
+size 402763943

pytorch_model_00023-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9bf0b3e4b78bb91c6c54cf7466da4f8e960ef75d7c8097e9477ec93ed23c98a5
+size 402763943

pytorch_model_00024-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fae8834fc4621493e84789d8d22d6a6cf30e231593642fc78ba2dfc740a7a85d
+size 402763943

pytorch_model_00025-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d3b83068eb4f581d5cb6c2b6007238c6ce5f901e2d45940a51c125be7dafcdf
+size 402763943

pytorch_model_00026-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8a200ef50432cf4717e079e19f810fcdb1a31e438780def80ec35ea3902ad56
+size 402763943

pytorch_model_00027-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac5531646ec9149a56d4e193ef745461ef9730d69f24df55738702a0675de09f
+size 402763943

pytorch_model_00028-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:112ffbcfaf5b50aadb0418c369318abe2507503a0b7e74960950c65fc4846faa
+size 402763943

pytorch_model_00029-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef47051c33ba7bbf5ca4aef33e9e888674559becb513462cee861a4c40ee0b1d
+size 402763943

pytorch_model_00030-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:695f43d15a990d6ddbe914d386672dfd34ed3e79072a01e6b9051ad4b58236aa
+size 402763943

pytorch_model_00031-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:918f6e4a9e38dbaceb8a860643ded8799c7968a28567c37f68b4bf15e61d6f9c
+size 402763943

pytorch_model_00032-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f58caf88f2f81f6e2208a3905f6d5c9d8edcc73c207493f9208a86e870dc362
+size 17595

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fa39cd4b1500feb205bcce3b9703a4373414cafe4970e0657b413f7ddd2a9d3
+size 14500438

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<unk>", "eos_token": "</s>", "bos_token": "<s>", "pad_token": "<pad>", "name_or_path": "bigscience/tokenizer", "special_tokens_map_file": null, "tokenizer_class": "BloomTokenizerFast"}