Upload folder using huggingface_hub

.gitattributes

@@ -41,3 +41,12 @@ flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-t4w0b1ci/san
 flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-vink1pbs/sandbox/local_flytekit/5c0420d4c4b3f4d23493c53a092687d0/00000 filter=lfs diff=lfs merge=lfs -text
 flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-zlncfcoi/sandbox/local_flytekit/6aee1fb3e9250ccde375fc50d36bc745/00000 filter=lfs diff=lfs merge=lfs -text
 flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/0677b767d0e8c7dc792f2848d0fb74db3a03419e89b7f05c7421fe066f1906eb filter=lfs diff=lfs merge=lfs -text
+flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-8de7272m/sandbox/local_flytekit/02c5eb893e5fca72c779fefe3c460180/00000 filter=lfs diff=lfs merge=lfs -text
+flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-a1_cbdba/sandbox/local_flytekit/8afda6c7b216b86972db929a5abb97cc/00000 filter=lfs diff=lfs merge=lfs -text
+flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-dm9cm4bb/sandbox/local_flytekit/e3bb72188c161dbec9f46a6a6d9a7422/00000 filter=lfs diff=lfs merge=lfs -text
+flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-fhkzem9t/sandbox/local_flytekit/fdd08c03cb76b0b4299abd8cc9351da6/00000 filter=lfs diff=lfs merge=lfs -text
+flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-hxi8bltz/sandbox/local_flytekit/a1558b3bcb7c887d2e3e691aaead4925/00000 filter=lfs diff=lfs merge=lfs -text
+flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-q15qjvhe/sandbox/local_flytekit/700144dc556a4fab7ea42e73c2243d86/00000 filter=lfs diff=lfs merge=lfs -text
+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

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "EleutherAI/pythia-1B-deduped",
+  "_name_or_path": "togethercomputer/RedPajama-INCITE-Base-3B-v1",
   "architectures": [
     "GPTNeoXForCausalLM"
   ],
@@ -7,20 +7,20 @@
   "classifier_dropout": 0.1,
   "eos_token_id": 0,
   "hidden_act": "gelu",
-  "hidden_size": 2048,
+  "hidden_size": 2560,
   "initializer_range": 0.02,
-  "intermediate_size": 8192,
+  "intermediate_size": 10240,
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 2048,
   "model_type": "gpt_neox",
-  "num_attention_heads": 8,
-  "num_hidden_layers": 16,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
   "rotary_emb_base": 10000,
-  "rotary_pct": 0.25,
+  "rotary_pct": 1.0,
   "tie_word_embeddings": false,
   "torch_dtype": "float16",
   "transformers_version": "4.29.2",
   "use_cache": true,
-  "use_parallel_residual": true,
-  "vocab_size": 50304
+  "use_parallel_residual": false,
+  "vocab_size": 50432
 }

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/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
+}

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-8de7272m/sandbox/local_flytekit/02c5eb893e5fca72c779fefe3c460180/00000

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-a1_cbdba/sandbox/local_flytekit/8afda6c7b216b86972db929a5abb97cc/00000

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-dm9cm4bb/sandbox/local_flytekit/e3bb72188c161dbec9f46a6a6d9a7422/00000

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-fhkzem9t/sandbox/local_flytekit/fdd08c03cb76b0b4299abd8cc9351da6/00000

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-htp2e655/sandbox/local_flytekit/inputs.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a421bfcdd8008c1ab1695d88ad84012ff0d43ffd4047456bb6c9104942a1de8f
+size 1839

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-hxi8bltz/sandbox/local_flytekit/a1558b3bcb7c887d2e3e691aaead4925/00000

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-q15qjvhe/sandbox/local_flytekit/700144dc556a4fab7ea42e73c2243d86/00000

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-q6jsg6o0/sandbox/local_flytekit/7386fa81441ede9aba39d83f5bb504bf/00000

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte-w81ns8fy/sandbox/local_flytekit/5f61f913333815551827fa3f4d3cc54a/00000

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/flyte_training_config.json

@@ -0,0 +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": ""}

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/generation_config.json

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/blobs/0204ed10c186a4c7c68f55dff8f26087a45898d6

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/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"
+}

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/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
+}

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/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|>"
+}

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/models--togethercomputer--RedPajama-INCITE-Base-3B-v1/refs/main

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/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
+}

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/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"
+}

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/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|>"
+}

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/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|>"
+}

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:747543e20635e50ec287e6ab96d89482a765b1459a8ca932e77b9d5033d1acd9
+size 5686106713

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/special_tokens_map.json

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

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/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|>"
+}

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/trainer_state.json

@@ -0,0 +1,31 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 6.666666666666667,
+  "global_step": 10,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 6.67,
+      "learning_rate": 2e-05,
+      "loss": 2.0296,
+      "step": 10
+    },
+    {
+      "epoch": 6.67,
+      "step": 10,
+      "total_flos": 4181427486720.0,
+      "train_loss": 2.029572677612305,
+      "train_runtime": 431.4795,
+      "train_samples_per_second": 2.967,
+      "train_steps_per_second": 0.023
+    }
+  ],
+  "max_steps": 10,
+  "num_train_epochs": 10,
+  "total_flos": 4181427486720.0,
+  "trial_name": null,
+  "trial_params": null
+}

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea233ffe22ae0f54efe9e7c78d909a4e8222d25cd04ce62ed9ddfb7cf3ee00a8
+size 5499

flyte_4dtz63w/local_flytekit/5e5d536b4c662b7a5699a1dfbf8a9c95/version.txt

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

flyte_training_config.json

@@ -1 +1 @@
-{"base_model": "EleutherAI/pythia-1B-deduped", "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": 8, "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": 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": ""}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3815356a26467dbb22ff4925dad4b33727d4c7c3d9816e9872711058d69016c5
-size 1285449053
+oid sha256:8dc592d4ce8dce386338bcd65743306043c0b55634bd1268470d3f6188fdfa76
+size 3169529757

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c85c668f7c6177cf1bc693b5aa289624c794d7783baa149a857a3634c259c98f
+oid sha256:60b5f769534832c130ff05e2a2bc6bc0decda18bdda584cc6d008f804bc1413b
 size 3899