Upload 8 files

.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

README.md

@@ -1,3 +1,54 @@
 ---
 license: bigscience-bloom-rail-1.0
+tags:
+- generated_from_trainer
+datasets:
+- squad_v2
+model-index:
+- name: debug_bloom_squad
+  results: []
 ---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# debug_bloom_squad
+
+This model is a fine-tuned version of [bigscience/bloom-560m](https://huggingface.co/bigscience/bloom-560m) on the squad_v2 dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 3e-05
+- train_batch_size: 6
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 2.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.24.0.dev0
+- Pytorch 1.12.1+cu102
+- Datasets 2.6.1
+- Tokenizers 0.13.1

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.0,
+    "train_loss": 3.5304283500368654,
+    "train_runtime": 15323.5402,
+    "train_samples": 131854,
+    "train_samples_per_second": 17.209,
+    "train_steps_per_second": 2.868
+}

config.json

@@ -0,0 +1,32 @@
+{
+  "_name_or_path": "bigscience/bloom-560m",
+  "apply_residual_connection_post_layernorm": false,
+  "architectures": [
+    "BloomForQuestionAnswering"
+  ],
+  "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": "4.24.0.dev0",
+  "unk_token_id": 0,
+  "use_cache": true,
+  "vocab_size": 250880
+}

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f6efc66e73f1fd69da4f436e48befb519fdff3fe18910850c1d41bd862293a5
+size 14500443

tokenizer_config.json

@@ -0,0 +1,11 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "name_or_path": "bigscience/bloom-560m",
+  "pad_token": "<pad>",
+  "padding_side": "left",
+  "special_tokens_map_file": null,
+  "tokenizer_class": "BloomTokenizer",
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.0,
+    "train_loss": 3.5304283500368654,
+    "train_runtime": 15323.5402,
+    "train_samples": 131854,
+    "train_samples_per_second": 17.209,
+    "train_steps_per_second": 2.868
+}

trainer_state.json

@@ -0,0 +1,547 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "global_step": 43952,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9665544230069168e-05,
+      "loss": 8.2605,
+      "step": 500
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 2.9324262832180562e-05,
+      "loss": 4.2073,
+      "step": 1000
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 2.8982981434291954e-05,
+      "loss": 4.0271,
+      "step": 1500
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 2.864170003640335e-05,
+      "loss": 3.8021,
+      "step": 2000
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 2.8300418638514747e-05,
+      "loss": 3.7889,
+      "step": 2500
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 2.795913724062614e-05,
+      "loss": 3.9639,
+      "step": 3000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 2.7617855842737534e-05,
+      "loss": 3.7584,
+      "step": 3500
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 2.7276574444848925e-05,
+      "loss": 3.9555,
+      "step": 4000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 2.6935975609756097e-05,
+      "loss": 3.9079,
+      "step": 4500
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 2.6594694211867496e-05,
+      "loss": 3.9956,
+      "step": 5000
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 2.6253412813978887e-05,
+      "loss": 3.1935,
+      "step": 5500
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 2.5912131416090282e-05,
+      "loss": 3.772,
+      "step": 6000
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 2.5572215143793228e-05,
+      "loss": 5.1608,
+      "step": 6500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 2.5230933745904623e-05,
+      "loss": 4.1891,
+      "step": 7000
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 2.4889652348016018e-05,
+      "loss": 3.9175,
+      "step": 7500
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 2.4548370950127413e-05,
+      "loss": 3.9636,
+      "step": 8000
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 2.4207089552238807e-05,
+      "loss": 4.1169,
+      "step": 8500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.38658081543502e-05,
+      "loss": 3.9023,
+      "step": 9000
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 2.3524526756461594e-05,
+      "loss": 3.8476,
+      "step": 9500
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.3183245358572992e-05,
+      "loss": 3.7756,
+      "step": 10000
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.2841963960684384e-05,
+      "loss": 4.0936,
+      "step": 10500
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 2.250068256279578e-05,
+      "loss": 3.7001,
+      "step": 11000
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 2.215940116490717e-05,
+      "loss": 4.269,
+      "step": 11500
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 2.1818119767018565e-05,
+      "loss": 3.8019,
+      "step": 12000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 2.147752093192574e-05,
+      "loss": 3.8787,
+      "step": 12500
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 2.113692209683291e-05,
+      "loss": 3.9497,
+      "step": 13000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 2.0795640698944304e-05,
+      "loss": 3.8454,
+      "step": 13500
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 2.0454359301055696e-05,
+      "loss": 3.824,
+      "step": 14000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.011307790316709e-05,
+      "loss": 3.9265,
+      "step": 14500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.977179650527849e-05,
+      "loss": 3.9985,
+      "step": 15000
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.943051510738988e-05,
+      "loss": 3.7808,
+      "step": 15500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.9089233709501275e-05,
+      "loss": 4.0853,
+      "step": 16000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.8747952311612667e-05,
+      "loss": 4.1285,
+      "step": 16500
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.840735347651984e-05,
+      "loss": 3.4963,
+      "step": 17000
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8066072078631237e-05,
+      "loss": 3.8048,
+      "step": 17500
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.772479068074263e-05,
+      "loss": 3.5536,
+      "step": 18000
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.7383509282854024e-05,
+      "loss": 3.6811,
+      "step": 18500
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.7042910447761196e-05,
+      "loss": 3.6341,
+      "step": 19000
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6701629049872587e-05,
+      "loss": 3.5169,
+      "step": 19500
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6360347651983986e-05,
+      "loss": 3.9027,
+      "step": 20000
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6019066254095377e-05,
+      "loss": 3.9672,
+      "step": 20500
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.567846741900255e-05,
+      "loss": 3.6631,
+      "step": 21000
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5337186021113944e-05,
+      "loss": 3.4308,
+      "step": 21500
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4995904623225337e-05,
+      "loss": 3.4298,
+      "step": 22000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.465462322533673e-05,
+      "loss": 3.5444,
+      "step": 22500
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.4313341827448125e-05,
+      "loss": 3.1415,
+      "step": 23000
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.397206042955952e-05,
+      "loss": 3.1462,
+      "step": 23500
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.3630779031670913e-05,
+      "loss": 3.0745,
+      "step": 24000
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.3289497633782308e-05,
+      "loss": 2.8795,
+      "step": 24500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.294889879868948e-05,
+      "loss": 2.9813,
+      "step": 25000
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.2607617400800874e-05,
+      "loss": 2.9452,
+      "step": 25500
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.2266336002912269e-05,
+      "loss": 3.1609,
+      "step": 26000
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.1925054605023662e-05,
+      "loss": 3.0651,
+      "step": 26500
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.1584455769930834e-05,
+      "loss": 3.1032,
+      "step": 27000
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.1243174372042229e-05,
+      "loss": 3.1298,
+      "step": 27500
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.0901892974153622e-05,
+      "loss": 3.033,
+      "step": 28000
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.0560611576265017e-05,
+      "loss": 3.0212,
+      "step": 28500
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.0220012741172187e-05,
+      "loss": 2.9057,
+      "step": 29000
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 9.878731343283582e-06,
+      "loss": 3.1331,
+      "step": 29500
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 9.537449945394977e-06,
+      "loss": 3.1259,
+      "step": 30000
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 9.19616854750637e-06,
+      "loss": 2.7155,
+      "step": 30500
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 8.855569712413542e-06,
+      "loss": 2.9427,
+      "step": 31000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 8.514288314524936e-06,
+      "loss": 2.9428,
+      "step": 31500
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 8.17300691663633e-06,
+      "loss": 3.1678,
+      "step": 32000
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 7.831725518747725e-06,
+      "loss": 3.4137,
+      "step": 32500
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 7.491126683654897e-06,
+      "loss": 2.9195,
+      "step": 33000
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 7.149845285766291e-06,
+      "loss": 3.3734,
+      "step": 33500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 6.808563887877685e-06,
+      "loss": 3.2046,
+      "step": 34000
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 6.467282489989079e-06,
+      "loss": 3.1923,
+      "step": 34500
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 6.126683654896251e-06,
+      "loss": 3.0387,
+      "step": 35000
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 5.785402257007645e-06,
+      "loss": 3.0879,
+      "step": 35500
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 5.444120859119039e-06,
+      "loss": 3.0943,
+      "step": 36000
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 5.102839461230433e-06,
+      "loss": 3.1384,
+      "step": 36500
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 4.762240626137604e-06,
+      "loss": 2.8415,
+      "step": 37000
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 4.420959228248999e-06,
+      "loss": 3.0139,
+      "step": 37500
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 4.079677830360393e-06,
+      "loss": 3.1161,
+      "step": 38000
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 3.7383964324717874e-06,
+      "loss": 3.2882,
+      "step": 38500
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 3.3977975973789587e-06,
+      "loss": 3.0338,
+      "step": 39000
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 3.056516199490353e-06,
+      "loss": 2.876,
+      "step": 39500
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 2.7152348016017477e-06,
+      "loss": 3.0888,
+      "step": 40000
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 2.3739534037131413e-06,
+      "loss": 3.1206,
+      "step": 40500
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 2.0326720058245358e-06,
+      "loss": 3.0954,
+      "step": 41000
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.6920731707317074e-06,
+      "loss": 3.4009,
+      "step": 41500
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.3507917728431015e-06,
+      "loss": 3.1574,
+      "step": 42000
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.009510374954496e-06,
+      "loss": 2.8075,
+      "step": 42500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 6.682289770658902e-07,
+      "loss": 3.1462,
+      "step": 43000
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 3.2763014197306156e-07,
+      "loss": 3.0345,
+      "step": 43500
+    },
+    {
+      "epoch": 2.0,
+      "step": 43952,
+      "total_flos": 1.836818419405271e+17,
+      "train_loss": 3.5304283500368654,
+      "train_runtime": 15323.5402,
+      "train_samples_per_second": 17.209,
+      "train_steps_per_second": 2.868
+    }
+  ],
+  "max_steps": 43952,
+  "num_train_epochs": 2,
+  "total_flos": 1.836818419405271e+17,
+  "trial_name": null,
+  "trial_params": null
+}