Push to HF hub

.gitattributes

@@ -31,3 +31,4 @@ 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 +1,10 @@
+checkpoint-15000/
+word_embeddings.pt
+eval_results.json
+checkpoint-5000/
+word_embeddings_layernorm.pt
+all_results.json
+train_results.json
+checkpoint-25000/
+checkpoint-20000/
+checkpoint-10000/

config.json

@@ -0,0 +1,38 @@
+{
+  "_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,
+  "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
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e22b52d74d8dd006f0e100efd02ae6610b06c773f0a6d445b1aca1437f94bde3
+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"}

trainer_state.json

@@ -0,0 +1,125 @@
+{
+  "best_metric": 2.763396739959717,
+  "best_model_checkpoint": "/users/zyong2/data/zyong2/bigscience/data/processed/024/bloom-560m_de_continual-pretrain_100000samples_-1vocab_original_bsz2/checkpoint-25000",
+  "epoch": 0.6869075422448139,
+  "global_step": 25000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07,
+      "learning_rate": 9e-05,
+      "loss": 3.7384,
+      "step": 2500
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 8e-05,
+      "loss": 3.3183,
+      "step": 5000
+    },
+    {
+      "epoch": 0.14,
+      "eval_loss": 3.2969818115234375,
+      "eval_runtime": 84.13,
+      "eval_samples_per_second": 9.557,
+      "eval_steps_per_second": 4.778,
+      "step": 5000
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 7e-05,
+      "loss": 3.1486,
+      "step": 7500
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 6e-05,
+      "loss": 3.0317,
+      "step": 10000
+    },
+    {
+      "epoch": 0.27,
+      "eval_loss": 3.0825796127319336,
+      "eval_runtime": 84.1046,
+      "eval_samples_per_second": 9.56,
+      "eval_steps_per_second": 4.78,
+      "step": 10000
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 5e-05,
+      "loss": 2.944,
+      "step": 12500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4e-05,
+      "loss": 2.8657,
+      "step": 15000
+    },
+    {
+      "epoch": 0.41,
+      "eval_loss": 2.939875841140747,
+      "eval_runtime": 84.1494,
+      "eval_samples_per_second": 9.554,
+      "eval_steps_per_second": 4.777,
+      "step": 15000
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 3e-05,
+      "loss": 2.8046,
+      "step": 17500
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 2e-05,
+      "loss": 2.7595,
+      "step": 20000
+    },
+    {
+      "epoch": 0.55,
+      "eval_loss": 2.832326650619507,
+      "eval_runtime": 84.1324,
+      "eval_samples_per_second": 9.556,
+      "eval_steps_per_second": 4.778,
+      "step": 20000
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.7111,
+      "step": 22500
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.0,
+      "loss": 2.6609,
+      "step": 25000
+    },
+    {
+      "epoch": 0.69,
+      "eval_loss": 2.763396739959717,
+      "eval_runtime": 84.0579,
+      "eval_samples_per_second": 9.565,
+      "eval_steps_per_second": 4.782,
+      "step": 25000
+    },
+    {
+      "epoch": 0.69,
+      "step": 25000,
+      "total_flos": 9.28706985984e+16,
+      "train_loss": 2.99828,
+      "train_runtime": 18720.6702,
+      "train_samples_per_second": 2.671,
+      "train_steps_per_second": 1.335
+    }
+  ],
+  "max_steps": 25000,
+  "num_train_epochs": 1,
+  "total_flos": 9.28706985984e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35dd254bf6bc58b8714de3206666b3590f57b135a3c9c40e116e1d2da9af8038
+size 3375