End of training

README.md

@@ -0,0 +1,81 @@
+---
+base_model: roneneldan/TinyStories-33M
+library_name: Distily
+tags:
+- generated_from_trainer
+model-index:
+- name: distily_bench_obj_cross_v2.9
+  results: []
+---
+
+# distily_bench_obj_cross_v2.9
+
+This student model is distilled from the teacher model [roneneldan/TinyStories-33M](https://huggingface.co/roneneldan/TinyStories-33M) using the dataset (unspecified).
+
+The [Distily](https://github.com/lapp0/distily) library was used for this distillation.
+
+It achieves the following results on the evaluation set:
+- eval_enwikippl: 324.7383
+- eval_frwikippl: 357916.6562
+- eval_zhwikippl: 16872236.0
+- eval_tinystoriesppl: 10.5821
+- eval_loss: 1.0786
+- eval_runtime: 6.4992
+- eval_samples_per_second: 76.933
+- eval_steps_per_second: 9.694
+
+<!-- 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.
+
+## 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:
+- distillation_objective: DistillationObjective(logits_loss_component=LossComponent(label=logits, weight=1, loss_fn=kl, layer_mapper=None, projector=None), hs_loss_component=LossComponent(label=hs, weight=0, loss_fn=None, layer_mapper=None, projector=None), attn_loss_component=LossComponent(label=attn, weight=0, loss_fn=None, layer_mapper=None, projector=None))
+- train_embeddings: True
+- learning_rate: 0.0004
+- train_batch_size: 1
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 1.0
+
+### Resource Usage
+Peak GPU Memory: 6.6062 GB
+
+### Eval-Phase Metrics
+| step | epoch | enwikippl | frwikippl | loss | runtime | samples_per_second | steps_per_second | tinystoriesppl | zhwikippl |
+| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
+| **teacher eval** |  | 169.9865 | 47377.9414 |  |  |  |  | 3.9789 | 4998.1294 |
+| 0 | 0 | 60092.5469 | 59534.5039 | 6.5732 | 6.5296 | 76.574 | 9.648 | 63003.9648 | 58113.1992 |
+| 5000 | 0.1010 | 324.7383 | 357916.6562 | 1.0786 | 6.4992 | 76.933 | 9.694 | 10.5821 | 16872236.0 |
+| 10000 | 0.2020 | 356.0557 | 413217.4688 | 1.1040 | 6.4959 | 76.971 | 9.698 | 12.1085 | 21984360.0 |
+| 15000 | 0.3030 | 355.4769 | 399032.125 | 1.1187 | 6.4929 | 77.008 | 9.703 | 12.1451 | 12460803.0 |
+| 20000 | 0.4040 | 334.4266 | 447667.6875 | 1.1281 | 6.5143 | 76.754 | 9.671 | 10.0140 | 25472600.0 |
+| 25000 | 0.5051 | 302.2480 | 303727.4688 | 1.0864 | 6.5193 | 76.696 | 9.664 | 10.7403 | 14773035.0 |
+| 30000 | 0.6061 | 286.7610 | 294875.0312 | 1.0858 | 6.4951 | 76.981 | 9.7 | 10.0074 | 11514604.0 |
+| 35000 | 0.7071 | 285.1770 | 263654.5938 | 1.0990 | 6.4992 | 76.933 | 9.694 | 10.3828 | 10835058.0 |
+| 40000 | 0.8081 | 279.8266 | 256096.3594 | 1.0910 | 6.5095 | 76.811 | 9.678 | 10.2142 | 11009884.0 |
+| 45000 | 0.9091 | 281.9699 | 274029.3125 | 1.0897 | 6.5441 | 76.405 | 9.627 | 10.2510 | 11984763.0 |
+| 49500 | 1.0 | 282.4946 | 274183.8125 | 1.0899 | 6.5603 | 76.216 | 9.603 | 10.2595 | 12058527.0 |
+
+### Framework versions
+- Distily 0.2.0
+- Transformers 4.44.0
+- Pytorch 2.3.0
+- Datasets 2.21.0

config.json

@@ -0,0 +1,62 @@
+{
+  "_name_or_path": "roneneldan/TinyStories-33M",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPTNeoForCausalLM"
+  ],
+  "attention_dropout": 0,
+  "attention_layers": [
+    "global",
+    "local",
+    "global",
+    "local"
+  ],
+  "attention_types": [
+    [
+      [
+        "global",
+        "local"
+      ],
+      2
+    ]
+  ],
+  "bos_token_id": 50256,
+  "classifier_dropout": 0.1,
+  "embed_dropout": 0,
+  "eos_token_id": 50256,
+  "gradient_checkpointing": false,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": null,
+  "layer_norm_epsilon": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neo",
+  "num_heads": 16,
+  "num_layers": 4,
+  "quantization_config": {
+    "_load_in_4bit": false,
+    "_load_in_8bit": true,
+    "bnb_4bit_compute_dtype": "float32",
+    "bnb_4bit_quant_storage": "uint8",
+    "bnb_4bit_quant_type": "fp4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": false,
+    "load_in_8bit": true,
+    "quant_method": "bitsandbytes"
+  },
+  "resid_dropout": 0,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.44.0",
+  "use_cache": true,
+  "vocab_size": 50257,
+  "window_size": 256
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "4.44.0"
+}

logs/copy_teacher_modules=__, dropout=0, learning_rate=0.0004, warmup_ratio=0, weight_decay=0/events.out.tfevents.1723945620.5f530b1cf724

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d924f66bcdfb2daa2d7375eb8420aa0267774daf02c3024a4685ce1e50326cf
+size 23672557

logs/copy_teacher_modules=__, dropout=0, learning_rate=0.0004, warmup_ratio=0, weight_decay=0/events.out.tfevents.1723949128.5f530b1cf724

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:475b00c4d2bb61b9ca3a194aa9291274dd55b954bb43bbada719413cdd0dc886
+size 312

logs/copy_teacher_modules=__, dropout=0, learning_rate=0.004, warmup_ratio=0, weight_decay=0/events.out.tfevents.1723944707.5f530b1cf724

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1a9e44a06cd845f331e9290f899a7d60cb011619a930299f206101012fd096a
+size 2365249

logs/copy_teacher_modules=__, dropout=0, learning_rate=0.004, warmup_ratio=0, weight_decay=0/events.out.tfevents.1723945180.5f530b1cf724

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1acfbac4a5544068cfe2b5ad9b37840af83a5d2b7f79ebc6da18dabd6765c44
+size 358566

logs/copy_teacher_modules=__, dropout=0, learning_rate=0.004, warmup_ratio=0, weight_decay=0/events.out.tfevents.1723945384.5f530b1cf724

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:039e8abddaddf5f57e5662902a7c41b5dfccfbedb45623ce9ebad308c64f4785
+size 535090

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72d926cfaaef9cc769840b758a5249acf11db15a4915dd82333b08d5b8e590ce
+size 137033984

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,22 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 2048,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dec9a785357983ea59aaacae23b2f7d4d6ac4274c758de9ced6c113fa663be0a
+size 1017948040