Upload folder using huggingface_hub

.gitattributes

@@ -41,3 +41,12 @@ flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-q15qjvhe/san
 flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-q6jsg6o0/sandbox/local_flytekit/7386fa81441ede9aba39d83f5bb504bf/00000 filter=lfs diff=lfs merge=lfs -text
 flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-w81ns8fy/sandbox/local_flytekit/5f61f913333815551827fa3f4d3cc54a/00000 filter=lfs diff=lfs merge=lfs -text
 flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/blobs/844708c42241de49775b872663ad58422a35b8e5a7d50b1f808408db8c1fb84b filter=lfs diff=lfs merge=lfs -text
+flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-6aqm1dej/sandbox/local_flytekit/9d4beefa976d280136f796b149e07526/00000 filter=lfs diff=lfs merge=lfs -text
+flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-7yc6imlm/sandbox/local_flytekit/963c360abe2edec91ada2515582ee92a/00000 filter=lfs diff=lfs merge=lfs -text
+flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-_x7eehjn/sandbox/local_flytekit/f621a93d05a475a754093f8944c82219/00000 filter=lfs diff=lfs merge=lfs -text
+flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-deze6p9p/sandbox/local_flytekit/399735041b6eef99517b2b1f194bd0c1/00000 filter=lfs diff=lfs merge=lfs -text
+flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-iyul5an6/sandbox/local_flytekit/f75cf20438a12d06b240bcc378551bc1/00000 filter=lfs diff=lfs merge=lfs -text
+flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-nyxd1kb4/sandbox/local_flytekit/6597b2a8ba65b3295dd362df112ba83f/00000 filter=lfs diff=lfs merge=lfs -text
+flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-tkhkqwit/sandbox/local_flytekit/bda70de7ed2fa38e66f14a88bdc1b202/00000 filter=lfs diff=lfs merge=lfs -text
+flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-xu53uvea/sandbox/local_flytekit/ec9d782ca167a7ca4f577c815dabbbb5/00000 filter=lfs diff=lfs merge=lfs -text
+flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/blobs/844708c42241de49775b872663ad58422a35b8e5a7d50b1f808408db8c1fb84b filter=lfs diff=lfs merge=lfs -text

flyte_training_config.json

@@ -1 +1 @@
-{"base_model": "togethercomputer/RedPajama-INCITE-Base-3B-v1", "data_path": "wikipedia", "data_name": "20220301.simple", "num_epochs": 1, "max_steps": 10, "learning_rate": 2e-05, "weight_decay": 0.02, "warmup_ratio": 0.03, "lr_scheduler_type": "cosine", "batch_size": 16, "micro_batch_size": 1, "val_set_size": 0, "group_by_length": false, "instruction_key": "instruction", "input_key": "input", "output_key": "output", "device_map": "auto", "cache_dir": null, "optim": "adamw_torch", "model_max_length": 512, "debug_mode": false, "debug_train_data_size": 1024, "wandb_project": ""}
+{"base_model": "togethercomputer/RedPajama-INCITE-Base-3B-v1", "data_path": "wikipedia", "data_name": "20220301.simple", "num_epochs": 3, "max_steps": -1, "learning_rate": 2e-05, "weight_decay": 0.02, "warmup_ratio": 0.03, "lr_scheduler_type": "cosine", "batch_size": 16, "micro_batch_size": 1, "val_set_size": 0, "group_by_length": false, "instruction_key": "instruction", "input_key": "input", "output_key": "output", "device_map": "auto", "cache_dir": null, "optim": "adamw_torch", "model_max_length": 512, "debug_mode": false, "debug_train_data_size": 1024, "wandb_project": ""}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "togethercomputer/RedPajama-INCITE-Base-3B-v1",
+  "architectures": [
+    "GPTNeoXForCausalLM"
+  ],
+  "bos_token_id": 0,
+  "classifier_dropout": 0.1,
+  "eos_token_id": 0,
+  "hidden_act": "gelu",
+  "hidden_size": 2560,
+  "initializer_range": 0.02,
+  "intermediate_size": 10240,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neox",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "rotary_emb_base": 10000,
+  "rotary_pct": 1.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.29.2",
+  "use_cache": true,
+  "use_parallel_residual": false,
+  "vocab_size": 50432
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-6aqm1dej/sandbox/local_flytekit/9d4beefa976d280136f796b149e07526/00000

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
+size 133886409

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-7yc6imlm/sandbox/local_flytekit/963c360abe2edec91ada2515582ee92a/00000

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
+size 133886409

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-_x7eehjn/sandbox/local_flytekit/f621a93d05a475a754093f8944c82219/00000

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
+size 133886409

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-deze6p9p/sandbox/local_flytekit/399735041b6eef99517b2b1f194bd0c1/00000

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
+size 133886409

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-iyul5an6/sandbox/local_flytekit/f75cf20438a12d06b240bcc378551bc1/00000

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
+size 133886409

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-nyxd1kb4/sandbox/local_flytekit/6597b2a8ba65b3295dd362df112ba83f/00000

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
+size 133886409

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-tkhkqwit/sandbox/local_flytekit/bda70de7ed2fa38e66f14a88bdc1b202/00000

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
+size 133886409

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-xu53uvea/sandbox/local_flytekit/ec9d782ca167a7ca4f577c815dabbbb5/00000

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
+size 133886409

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte-zulxylvx/sandbox/local_flytekit/inputs.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:090103b72486465b8c3311d5732f282a73657356b8458b952e89aae56d93540f
+size 1839

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/flyte_training_config.json

@@ -0,0 +1 @@
+{"base_model": "togethercomputer/RedPajama-INCITE-Base-3B-v1", "data_path": "wikipedia", "data_name": "20220301.simple", "num_epochs": 3, "max_steps": -1, "learning_rate": 2e-05, "weight_decay": 0.02, "warmup_ratio": 0.03, "lr_scheduler_type": "cosine", "batch_size": 16, "micro_batch_size": 1, "val_set_size": 0, "group_by_length": false, "instruction_key": "instruction", "input_key": "input", "output_key": "output", "device_map": "auto", "cache_dir": null, "optim": "adamw_torch", "model_max_length": 512, "debug_mode": false, "debug_train_data_size": 1024, "wandb_project": ""}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "transformers_version": "4.29.2"
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/blobs/0204ed10c186a4c7c68f55dff8f26087a45898d6

@@ -0,0 +1,5 @@
+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/blobs/344b8478e86f0cb7933a50ef1fff9127ed27266c

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "transformers_version": "4.28.1"
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/blobs/7cb72eb7357994a07d5c54ffdeef99dec84e21ae

@@ -0,0 +1,25 @@
+{
+  "_name_or_path": "rp_3b_800b",
+  "architectures": [
+    "GPTNeoXForCausalLM"
+  ],
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "hidden_act": "gelu",
+  "hidden_size": 2560,
+  "initializer_range": 0.02,
+  "intermediate_size": 10240,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neox",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "rotary_emb_base": 10000,
+  "rotary_pct": 1.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.28.1",
+  "use_cache": true,
+  "use_parallel_residual": false,
+  "vocab_size": 50432
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/blobs/844708c42241de49775b872663ad58422a35b8e5a7d50b1f808408db8c1fb84b

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:844708c42241de49775b872663ad58422a35b8e5a7d50b1f808408db8c1fb84b
+size 5686113497

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/blobs/af28df4f9190240e1c3fb77a27b6f3d9bdd41c2d

@@ -0,0 +1,9 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 2048,
+  "tokenizer_class": "GPTNeoXTokenizer",
+  "unk_token": "<|endoftext|>"
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/refs/main

@@ -0,0 +1 @@
+094fbdd0c911feb485ce55de1952ab2e75277e1e

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/snapshots/094fbdd0c911feb485ce55de1952ab2e75277e1e/config.json

@@ -0,0 +1,25 @@
+{
+  "_name_or_path": "rp_3b_800b",
+  "architectures": [
+    "GPTNeoXForCausalLM"
+  ],
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "hidden_act": "gelu",
+  "hidden_size": 2560,
+  "initializer_range": 0.02,
+  "intermediate_size": 10240,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neox",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "rotary_emb_base": 10000,
+  "rotary_pct": 1.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.28.1",
+  "use_cache": true,
+  "use_parallel_residual": false,
+  "vocab_size": 50432
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/snapshots/094fbdd0c911feb485ce55de1952ab2e75277e1e/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "transformers_version": "4.28.1"
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/snapshots/094fbdd0c911feb485ce55de1952ab2e75277e1e/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:844708c42241de49775b872663ad58422a35b8e5a7d50b1f808408db8c1fb84b
+size 5686113497

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/snapshots/094fbdd0c911feb485ce55de1952ab2e75277e1e/special_tokens_map.json

@@ -0,0 +1,5 @@
+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/snapshots/094fbdd0c911feb485ce55de1952ab2e75277e1e/tokenizer_config.json

@@ -0,0 +1,9 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 2048,
+  "tokenizer_class": "GPTNeoXTokenizer",
+  "unk_token": "<|endoftext|>"
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b572a401466eb4a6f0fb1935eb8b61a8ba0cc87888c5005b17220c8018f856a5
+size 5686106713

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "pad_token": "[PAD]",
+  "unk_token": "<|endoftext|>"
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/tokenizer_config.json

@@ -0,0 +1,11 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "padding_side": "right",
+  "tokenizer_class": "GPTNeoXTokenizer",
+  "unk_token": "<|endoftext|>"
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/trainer_state.json

@@ -0,0 +1,25 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "global_step": 3,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 2.0,
+      "step": 3,
+      "total_flos": 1236019445760.0,
+      "train_loss": 2.0709635416666665,
+      "train_runtime": 132.7651,
+      "train_samples_per_second": 4.271,
+      "train_steps_per_second": 0.023
+    }
+  ],
+  "max_steps": 3,
+  "num_train_epochs": 3,
+  "total_flos": 1236019445760.0,
+  "trial_name": null,
+  "trial_params": null
+}

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a779e63cb4d0789bd1182a9d2ff733fc92eb45d8ba4ea8762b6e0e264e86f43
+size 5563

flytexhkrcgxw/local_flytekit/a0bdcc43a58af4ab83094e2ad7e965a6/version.txt

@@ -0,0 +1 @@
+1

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8dc592d4ce8dce386338bcd65743306043c0b55634bd1268470d3f6188fdfa76
+oid sha256:5cf3264e3ed851419a0cd378b83a70dda4d7298add89b7b5d844e0cd890beb8f
 size 3169529757

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:60b5f769534832c130ff05e2a2bc6bc0decda18bdda584cc6d008f804bc1413b
+oid sha256:ee731ce9f4fb31d73fecfa97d3439645a74d7340661828ddc1b5632d1d41799e
 size 3899