Upload experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e with huggingface_hub

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/configs.json

@@ -0,0 +1,26 @@
+{
+    "output_dir": "./results",
+    "evaluation_strategy": "no",
+    "do_eval": false,
+    "eval_steps": 0,
+    "log_level": "info",
+    "logging_first_step": true,
+    "logging_steps": 5,
+    "logging_dir": "./logs",
+    "save_steps": 150,
+    "save_total_limit": 1,
+    "num_train_epochs": 10,
+    "per_device_train_batch_size": 6,
+    "optim": "adamw_torch",
+    "gradient_accumulation_steps": 64,
+    "dataloader_drop_last": true,
+    "warmup_steps": 500,
+    "weight_decay": 0.05,
+    "learning_rate": 5e-05,
+    "deepspeed": "deepspeed.json",
+    "local_rank": 0,
+    "total_gpus": 7,
+    "v_cpus": 128,
+    "total_memory_in_gb": 515836.75,
+    "dataset_limit": 0
+}

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/deepspeed.json

@@ -0,0 +1,48 @@
+{
+    "optimizer": {
+        "type": "AdamW",
+        "params": {
+            "lr": "auto",
+            "betas": "auto",
+            "eps": "auto",
+            "weight_decay": "auto"
+        }
+    },
+
+    "scheduler": {
+        "type": "WarmupLR",
+        "params": {
+            "warmup_min_lr": "auto",
+            "warmup_max_lr": "auto",
+            "warmup_num_steps": "auto"
+        }
+    },
+
+    "zero_optimization": {
+        "stage": 2,
+        "offload_optimizer": {
+            "device": "cpu",
+            "pin_memory": true
+        },
+        "allgather_partitions": true,
+        "allgather_bucket_size": 5e8,
+        "overlap_comm": true,
+        "reduce_scatter": true,
+        "reduce_bucket_size": 5e8,
+        "contiguous_gradients": true
+    },
+
+    "tensorboard": {
+        "enabled": true,
+        "output_path": "logs/",
+        "job_name": "train_neo"
+    },
+
+    "zero_allow_untested_optimizer": true,
+    "gradient_accumulation_steps": "auto",
+    "gradient_clipping": "auto",
+    "steps_per_print": 2000,
+    "train_batch_size": "auto",
+    "train_micro_batch_size_per_gpu": "auto",
+    "wall_clock_breakdown": false
+}

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/final_checkpoint/config.json

@@ -0,0 +1,54 @@
+{
+  "_name_or_path": "EleutherAI/gpt-neo-125M",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPTNeoForCausalLM"
+  ],
+  "attention_dropout": 0,
+  "attention_layers": [
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local"
+  ],
+  "attention_types": [
+    [
+      [
+        "global",
+        "local"
+      ],
+      6
+    ]
+  ],
+  "bos_token_id": 50256,
+  "embed_dropout": 0,
+  "eos_token_id": 50256,
+  "gradient_checkpointing": false,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": null,
+  "layer_norm_epsilon": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neo",
+  "num_heads": 12,
+  "num_layers": 12,
+  "resid_dropout": 0,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.24.0",
+  "use_cache": false,
+  "vocab_size": 50257,
+  "window_size": 256
+}

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/final_checkpoint/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8d5b471474c47702d4ca6d2965b4e309fb1d8622ee99e2599b04df9a24c9a88
+size 551154684

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/final_checkpoint/tokenizer/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/final_checkpoint/tokenizer/tokenizer_config.json

@@ -0,0 +1,34 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "bos_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "errors": "replace",
+  "model_max_length": 2048,
+  "name_or_path": "EleutherAI/gpt-neo-125M",
+  "pad_token": null,
+  "special_tokens_map_file": null,
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/1676906944.3890443/events.out.tfevents.1676906944.92030dbdb443.1285.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1aeb82fce1a3531af1cc40e914e93a6535d584dbc5e012032ad8264e7bab31f3
+size 5418

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/1676907104.787101/events.out.tfevents.1676907104.92030dbdb443.4422.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:387570a52fb35564bb6e129b8a88eb14a1a8d4c38ce50045e5a2c014b8ca89bb
+size 5418

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/1676907226.512263/events.out.tfevents.1676907226.92030dbdb443.6824.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8052c039bf94c60697bdb30d7362e29f2b6ea133c75d8bbf47597842f58c2026
+size 5418

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/1676907299.4790792/events.out.tfevents.1676907299.92030dbdb443.8606.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5487ad9b95f43319991e232de19cb783259a76a2dcd8f0e7ffcfd471359ffed
+size 5418

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/1676907447.1754365/events.out.tfevents.1676907447.92030dbdb443.10872.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4d1eb74a76b442fe66fdf07af462cb6058fad06531cc0861fb4da1a605e5e09
+size 5418

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/1676907758.8054497/events.out.tfevents.1676907758.92030dbdb443.13338.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a099c71fe4708aba1939b3be87d7b7810786bf23422f71c664faec15e8af469
+size 5418

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/1676907877.724431/events.out.tfevents.1676907877.92030dbdb443.15502.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f777b71d52a4319d934ad4cc57cb64a9e4ec505f83aa34cc3e0859693be78b8
+size 5418

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/1676912135.7173784/events.out.tfevents.1676912135.92030dbdb443.24388.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0ccca4bd5e17e914b1030b6fd9b8d007b7dd76d60534cbd308e35c2048c4092
+size 5418

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/1676912469.7910182/events.out.tfevents.1676912469.92030dbdb443.27348.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5610589b781fd2cb75231262339033b442a269404a4c95d583287bc055225ccb
+size 5418

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/1676912788.2396843/events.out.tfevents.1676912788.92030dbdb443.31458.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45f24dc75029c61d52f4c12cf97475650c2f2e2345ab6028547da56cf116307b
+size 5418

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/events.out.tfevents.1676906944.92030dbdb443.1285.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfa820831dcaf226451981d058fcdef53323284c33705c3b9d2acbfda75a32a2
+size 4202

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/events.out.tfevents.1676907104.92030dbdb443.4422.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d4e6ce25c665fdf72b3bfb63c11968b134358705147ee3f2cdb7afdb0b786a0
+size 4202

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/events.out.tfevents.1676907226.92030dbdb443.6824.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26b9c7a41f4cef0a57f1be1c6b2bb381c315585067e9d60933bc164c5fb85b62
+size 4048

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/events.out.tfevents.1676907299.92030dbdb443.8606.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32c0250ee08d1ca3771df078f3e508aa4228d6828d158be4ec3dc80a743313f5
+size 4202

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/events.out.tfevents.1676907447.92030dbdb443.10872.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da0440bcbb89df1012777d26432212a91254ea1df455b59dadc43b03314e6dd0
+size 4510

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/events.out.tfevents.1676907758.92030dbdb443.13338.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b834cde31d45d472b1d459aca16ef6a7faecc2063d2e4748fe5de39278d38735
+size 4202

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/events.out.tfevents.1676907877.92030dbdb443.15502.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a7371a441adf6a51452b9095069739593f6d41a1480bcfa24e71f80fdc2d4a9
+size 5742

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/events.out.tfevents.1676912135.92030dbdb443.24388.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61d840519037715ecca949eed6a1e87dd4f302b3b0a9465a0cd5721f0e0547aa
+size 4204

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/events.out.tfevents.1676912469.92030dbdb443.27348.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57697a9610bc3de732e244856fd5a07a7b817bdb5f8d7c9ec082309198a6f5d2
+size 4205

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/events.out.tfevents.1676912788.92030dbdb443.31458.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:366c6d481427752554410d58bc022d114db3d7392c22f1bc5d8eb4dbd9739f00
+size 17985

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/train_neo/events.out.tfevents.1676906885.92030dbdb443.1285.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5af3ae1657e58d602f13c0a85252f70bc2d5b74ac41b96068bd07be2ef6519
+size 572

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/train_neo/events.out.tfevents.1676907088.92030dbdb443.4422.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e9b6244eadfe76350454af66f7188069886e4813ec9693c67dfd85e83e2fe75
+size 155

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/train_neo/events.out.tfevents.1676907209.92030dbdb443.6824.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53787d2d30a5b884079de2726fed3ffa5a9a2df6a8d69319b168f049eaf8b512
+size 40

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/train_neo/events.out.tfevents.1676907283.92030dbdb443.8606.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b2d562fe78e1d79063987b769e15102b8bf8ac3241cb9132c412629ba1f007c
+size 509

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/train_neo/events.out.tfevents.1676907430.92030dbdb443.10872.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86ffafe6c17af68fa55334caa67139556cab03dacc9be0d20712112a83476fa6
+size 1453

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/train_neo/events.out.tfevents.1676907742.92030dbdb443.13338.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc4ad867eae7c1016f1458eb23b36910771b0cf335cf30eb2c4522a368f164bf
+size 155

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/train_neo/events.out.tfevents.1676907861.92030dbdb443.15502.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19a84bb96b62dc925994fda31e9d919464a476d90d72d353d4e0d61725029b43
+size 6144

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/train_neo/events.out.tfevents.1676912119.92030dbdb443.24388.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48437de406d942b8b1caef30eda5b0bd36e4b9b83f12abae367a1cf8b981af19
+size 273

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/train_neo/events.out.tfevents.1676912453.92030dbdb443.27348.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a73c21c50d8cc55cab95bd16c5a4cd0cdab239ca78fb21391d291ae5268de8a
+size 155

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/logs/train_neo/events.out.tfevents.1676912772.92030dbdb443.31458.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6d0ad6e97ab3588d9615ea7b10c598fd8bc7118ed2e4746411447f81f44e06b
+size 52224

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/trainer_final_checkpoint/config.json

@@ -0,0 +1,54 @@
+{
+  "_name_or_path": "EleutherAI/gpt-neo-125M",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPTNeoForCausalLM"
+  ],
+  "attention_dropout": 0,
+  "attention_layers": [
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local"
+  ],
+  "attention_types": [
+    [
+      [
+        "global",
+        "local"
+      ],
+      6
+    ]
+  ],
+  "bos_token_id": 50256,
+  "embed_dropout": 0,
+  "eos_token_id": 50256,
+  "gradient_checkpointing": false,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": null,
+  "layer_norm_epsilon": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neo",
+  "num_heads": 12,
+  "num_layers": 12,
+  "resid_dropout": 0,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.24.0",
+  "use_cache": false,
+  "vocab_size": 50257,
+  "window_size": 256
+}

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/trainer_final_checkpoint/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8d5b471474c47702d4ca6d2965b4e309fb1d8622ee99e2599b04df9a24c9a88
+size 551154684

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/trainer_final_checkpoint/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/trainer_final_checkpoint/tokenizer_config.json

@@ -0,0 +1,34 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "bos_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "errors": "replace",
+  "model_max_length": 2048,
+  "name_or_path": "EleutherAI/gpt-neo-125M",
+  "pad_token": null,
+  "special_tokens_map_file": null,
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

experiments/2023-02-21-b0010c97cb1f06debca911602ea05b6ff85a8270fb9487d27b3d52eb4eb29e9e/trainer_final_checkpoint/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8cbfab6ac4e393b1fc2c734b6e39afa0448307ce7bfa29da66852dcf0f070f0
+size 4411