Push to HF hub

.gitattributes

@@ -30,3 +30,4 @@ 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.json filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,3 @@
+checkpoint-*/
+*/pilot_*/
+pilot_*/

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 0.4,
+    "eval_loss": 0.5791521668434143,
+    "eval_runtime": 501.8375,
+    "eval_samples": 1000,
+    "eval_samples_per_second": 9.401,
+    "eval_steps_per_second": 4.701,
+    "perplexity": 1.784524809889192,
+    "train_loss": 0.6467895458984375,
+    "train_runtime": 73141.1233,
+    "train_samples": 100000,
+    "train_samples_per_second": 2.734,
+    "train_steps_per_second": 0.342
+}

config.json

@@ -0,0 +1,39 @@
+{
+  "_name_or_path": "bigscience/bloom-560m",
+  "adapters": {
+    "adapters": {},
+    "config_map": {},
+    "fusion_config_map": {},
+    "fusions": {}
+  },
+  "apply_residual_connection_post_layernorm": false,
+  "architectures": [
+    "BloomForCausalLM"
+  ],
+  "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,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "masked_softmax_fusion": true,
+  "model_type": "bloom",
+  "n_head": 16,
+  "n_inner": null,
+  "n_layer": 24,
+  "offset_alibi": 100,
+  "pad_token_id": 3,
+  "pretraining_tp": 1,
+  "seq_length": 2048,
+  "skip_bias_add": true,
+  "skip_bias_add_qkv": false,
+  "slow_but_exact": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.20.0.dev0",
+  "unk_token_id": 0,
+  "use_cache": true,
+  "vocab_size": 250880
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.4,
+    "eval_loss": 0.5791521668434143,
+    "eval_runtime": 501.8375,
+    "eval_samples": 1000,
+    "eval_samples_per_second": 9.401,
+    "eval_steps_per_second": 4.701,
+    "perplexity": 1.784524809889192
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78a908a7d88ee13e8c406978ef30e18fd09db139791dc40bf07539ff854140cc
+size 2236955191

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 sha256:8f6efc66e73f1fd69da4f436e48befb519fdff3fe18910850c1d41bd862293a5
+size 14500443

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "<pad>", "add_prefix_space": false, "name_or_path": "bigscience/bloom-560m", "special_tokens_map_file": null, "padding_side": "left", "tokenizer_class": "BloomTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.4,
+    "train_loss": 0.6467895458984375,
+    "train_runtime": 73141.1233,
+    "train_samples": 100000,
+    "train_samples_per_second": 2.734,
+    "train_steps_per_second": 0.342
+}

trainer_state.json

@@ -0,0 +1,125 @@
+{
+  "best_metric": 0.5791521668434143,
+  "best_model_checkpoint": "/home/khalid/Documents/github_rep/bigscience/data/processed/21/bloom-560m_my_continual-pretrain_100000samples_-1vocab_original/checkpoint-25000",
+  "epoch": 0.39669788679035706,
+  "global_step": 25000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 9e-05,
+      "loss": 0.8854,
+      "step": 2500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 8e-05,
+      "loss": 0.7087,
+      "step": 5000
+    },
+    {
+      "epoch": 0.08,
+      "eval_loss": 0.7005436420440674,
+      "eval_runtime": 502.012,
+      "eval_samples_per_second": 9.398,
+      "eval_steps_per_second": 4.699,
+      "step": 5000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 7e-05,
+      "loss": 0.6665,
+      "step": 7500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 6e-05,
+      "loss": 0.6442,
+      "step": 10000
+    },
+    {
+      "epoch": 0.16,
+      "eval_loss": 0.6457962393760681,
+      "eval_runtime": 502.0206,
+      "eval_samples_per_second": 9.398,
+      "eval_steps_per_second": 4.699,
+      "step": 10000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 5e-05,
+      "loss": 0.6229,
+      "step": 12500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4e-05,
+      "loss": 0.6083,
+      "step": 15000
+    },
+    {
+      "epoch": 0.24,
+      "eval_loss": 0.6155888438224792,
+      "eval_runtime": 501.1781,
+      "eval_samples_per_second": 9.414,
+      "eval_steps_per_second": 4.707,
+      "step": 15000
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 3e-05,
+      "loss": 0.5988,
+      "step": 17500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 2e-05,
+      "loss": 0.5859,
+      "step": 20000
+    },
+    {
+      "epoch": 0.32,
+      "eval_loss": 0.5931491255760193,
+      "eval_runtime": 501.6202,
+      "eval_samples_per_second": 9.406,
+      "eval_steps_per_second": 4.703,
+      "step": 20000
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 0.5767,
+      "step": 22500
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0,
+      "loss": 0.5705,
+      "step": 25000
+    },
+    {
+      "epoch": 0.4,
+      "eval_loss": 0.5791521668434143,
+      "eval_runtime": 501.9862,
+      "eval_samples_per_second": 9.399,
+      "eval_steps_per_second": 4.699,
+      "step": 25000
+    },
+    {
+      "epoch": 0.4,
+      "step": 25000,
+      "total_flos": 3.714827943936e+17,
+      "train_loss": 0.6467895458984375,
+      "train_runtime": 73141.1233,
+      "train_samples_per_second": 2.734,
+      "train_steps_per_second": 0.342
+    }
+  ],
+  "max_steps": 25000,
+  "num_train_epochs": 1,
+  "total_flos": 3.714827943936e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd62139b92644490829901c392f1cd9b0aca6b39e504ec82be776fa751e92b3c
+size 3375

wikiann-my-results.txt

@@ -0,0 +1,8 @@
+==================================================
+Results
+==================================================
+Model: /home/khalid/Documents/github_rep/bigscience/data/processed/22/bloom-560m_my_continual-pretrain_100000samples_-1vocab_original/
+[0.0, 0.04794520547945205, 0.09756097560975609, 0.043343653250773995, 0.02867383512544803, 0.02476780185758514, 0.007220216606498194, 0.04294478527607362, 0.02958579881656805, 0.051660516605166046]
+3.74
+2.57
+==================================================

word_embeddings.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8896d95df461c567c41f6662a32db1d29044512d19dfc1f489e09e1d1b189568
+size 1027605867

word_embeddings_layernorm.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fc24defb3828704fcaf2edc1284529c5e17756ff4694a74aa87af914facffca
+size 9703