Super-squash branch 'main' using huggingface_hub

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz 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

README.md

@@ -0,0 +1,26 @@
+---
+license: apache-2.0
+base_model: Qwen/Qwen2-1.5B
+metrics:
+- accuracy
+datasets:
+- BEE-spoke-data/stepbasin-books
+language:
+- en
+---
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/pszemraj/long-generation-tests/runs/ethp25f9)
+# Qwen2-1.5B-stepbasin-books
+
+> [!IMPORTANT]  
+> this was finetuned at 16384 context length
+
+This is an experiment on long context text **generation** (i.e. 6k+ tokens generated) to evaluate if/when generation breaks down, etc. As such, all the data on which this model has been fine-tuned are full-length books.
+## Details
+
+This model is a fine-tuned version of [Qwen/Qwen2-1.5B](https://huggingface.co/Qwen/Qwen2-1.5B) on https://github.com/stepbasin/books/tree/master/books
+
+It achieves the following results on the evaluation set:
+- Loss: 2.8110
+- Accuracy: 0.4298
+- Num Input Tokens Seen: 44040192

added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 2.9899888765294773,
+    "eval_accuracy": 0.4297979192055684,
+    "eval_loss": 2.810983180999756,
+    "eval_runtime": 19.3217,
+    "eval_samples": 29,
+    "eval_samples_per_second": 1.501,
+    "eval_steps_per_second": 1.501,
+    "num_input_tokens_seen": 44040192,
+    "perplexity": 16.62625680536472,
+    "total_flos": 3.462459117703004e+17,
+    "train_loss": 2.7097946518943425,
+    "train_runtime": 6795.1554,
+    "train_samples": 899,
+    "train_samples_per_second": 0.397,
+    "train_steps_per_second": 0.012
+}

config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "Qwen/Qwen2-1.5B",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 131072,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 1000000.0,
+  "sliding_window": 131072,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.42.4",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 2.9899888765294773,
+    "eval_accuracy": 0.4297979192055684,
+    "eval_loss": 2.810983180999756,
+    "eval_runtime": 19.3217,
+    "eval_samples": 29,
+    "eval_samples_per_second": 1.501,
+    "eval_steps_per_second": 1.501,
+    "num_input_tokens_seen": 44040192,
+    "perplexity": 16.62625680536472
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 8192,
+  "do_sample": true,
+  "repetition_penalty": 1.04,
+  "renormalize_logits": true,
+  "no_repeat_ngram_size": 4,
+  "epsilon_cutoff": 1e-5
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f0ee7d4553edd7103ba253a2eb1ac64e606af2fa72d726b8aaabd8461dd847b
+size 3087467144

special_tokens_map.json

@@ -0,0 +1,20 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,42 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

train_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 2.9899888765294773,
+    "num_input_tokens_seen": 44040192,
+    "total_flos": 3.462459117703004e+17,
+    "train_loss": 2.7097946518943425,
+    "train_runtime": 6795.1554,
+    "train_samples": 899,
+    "train_samples_per_second": 0.397,
+    "train_steps_per_second": 0.012
+}

trainer_state.json

@@ -0,0 +1,297 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.9899888765294773,
+  "eval_steps": 300.0,
+  "global_step": 84,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1067853170189099,
+      "grad_norm": 0.5859375,
+      "learning_rate": 1.8e-05,
+      "loss": 2.8121,
+      "num_input_tokens_seen": 1572864,
+      "step": 3
+    },
+    {
+      "epoch": 0.2135706340378198,
+      "grad_norm": 0.69921875,
+      "learning_rate": 2.9988140958409528e-05,
+      "loss": 2.8211,
+      "num_input_tokens_seen": 3145728,
+      "step": 6
+    },
+    {
+      "epoch": 0.3203559510567297,
+      "grad_norm": 0.796875,
+      "learning_rate": 2.9810630129045003e-05,
+      "loss": 2.7286,
+      "num_input_tokens_seen": 4718592,
+      "step": 9
+    },
+    {
+      "epoch": 0.4271412680756396,
+      "grad_norm": 0.71875,
+      "learning_rate": 2.9422573564911305e-05,
+      "loss": 2.7688,
+      "num_input_tokens_seen": 6291456,
+      "step": 12
+    },
+    {
+      "epoch": 0.5339265850945495,
+      "grad_norm": 0.58984375,
+      "learning_rate": 2.88294878209231e-05,
+      "loss": 2.756,
+      "num_input_tokens_seen": 7864320,
+      "step": 15
+    },
+    {
+      "epoch": 0.6407119021134594,
+      "grad_norm": 0.54296875,
+      "learning_rate": 2.8039804116593743e-05,
+      "loss": 2.7252,
+      "num_input_tokens_seen": 9437184,
+      "step": 18
+    },
+    {
+      "epoch": 0.7474972191323693,
+      "grad_norm": 0.54296875,
+      "learning_rate": 2.7064748479061476e-05,
+      "loss": 2.7188,
+      "num_input_tokens_seen": 11010048,
+      "step": 21
+    },
+    {
+      "epoch": 0.8542825361512792,
+      "grad_norm": 0.484375,
+      "learning_rate": 2.5918182155542415e-05,
+      "loss": 2.7574,
+      "num_input_tokens_seen": 12582912,
+      "step": 24
+    },
+    {
+      "epoch": 0.9610678531701891,
+      "grad_norm": 0.5,
+      "learning_rate": 2.4616404563883302e-05,
+      "loss": 2.7792,
+      "num_input_tokens_seen": 14155776,
+      "step": 27
+    },
+    {
+      "epoch": 0.996662958843159,
+      "eval_accuracy": 0.4287265815910942,
+      "eval_loss": 2.818293809890747,
+      "eval_runtime": 19.7558,
+      "eval_samples_per_second": 1.468,
+      "eval_steps_per_second": 1.468,
+      "num_input_tokens_seen": 14729216,
+      "step": 28
+    },
+    {
+      "epoch": 1.067853170189099,
+      "grad_norm": 0.50390625,
+      "learning_rate": 2.3177921582440015e-05,
+      "loss": 2.6953,
+      "num_input_tokens_seen": 15728640,
+      "step": 30
+    },
+    {
+      "epoch": 1.174638487208009,
+      "grad_norm": 0.462890625,
+      "learning_rate": 2.162318247323868e-05,
+      "loss": 2.7098,
+      "num_input_tokens_seen": 17301504,
+      "step": 33
+    },
+    {
+      "epoch": 1.2814238042269188,
+      "grad_norm": 0.451171875,
+      "learning_rate": 1.997428917828102e-05,
+      "loss": 2.6726,
+      "num_input_tokens_seen": 18874368,
+      "step": 36
+    },
+    {
+      "epoch": 1.3882091212458287,
+      "grad_norm": 0.462890625,
+      "learning_rate": 1.825468212159477e-05,
+      "loss": 2.6655,
+      "num_input_tokens_seen": 20447232,
+      "step": 39
+    },
+    {
+      "epoch": 1.4949944382647387,
+      "grad_norm": 0.455078125,
+      "learning_rate": 1.6488806983620927e-05,
+      "loss": 2.7292,
+      "num_input_tokens_seen": 22020096,
+      "step": 42
+    },
+    {
+      "epoch": 1.6017797552836486,
+      "grad_norm": 0.4453125,
+      "learning_rate": 1.4701767185023948e-05,
+      "loss": 2.6312,
+      "num_input_tokens_seen": 23592960,
+      "step": 45
+    },
+    {
+      "epoch": 1.7085650723025583,
+      "grad_norm": 0.43359375,
+      "learning_rate": 1.2918967020163978e-05,
+      "loss": 2.7176,
+      "num_input_tokens_seen": 25165824,
+      "step": 48
+    },
+    {
+      "epoch": 1.8153503893214684,
+      "grad_norm": 0.4453125,
+      "learning_rate": 1.116575051339288e-05,
+      "loss": 2.7234,
+      "num_input_tokens_seen": 26738688,
+      "step": 51
+    },
+    {
+      "epoch": 1.9221357063403781,
+      "grad_norm": 0.4375,
+      "learning_rate": 9.467041132139884e-06,
+      "loss": 2.6971,
+      "num_input_tokens_seen": 28311552,
+      "step": 54
+    },
+    {
+      "epoch": 1.993325917686318,
+      "eval_accuracy": 0.4297116228554831,
+      "eval_loss": 2.8111748695373535,
+      "eval_runtime": 19.5837,
+      "eval_samples_per_second": 1.481,
+      "eval_steps_per_second": 1.481,
+      "num_input_tokens_seen": 29458432,
+      "step": 56
+    },
+    {
+      "epoch": 2.0289210233592883,
+      "grad_norm": 0.41796875,
+      "learning_rate": 7.846987478572411e-06,
+      "loss": 2.6623,
+      "num_input_tokens_seen": 29884416,
+      "step": 57
+    },
+    {
+      "epoch": 2.135706340378198,
+      "grad_norm": 0.396484375,
+      "learning_rate": 6.328619996627272e-06,
+      "loss": 2.6211,
+      "num_input_tokens_seen": 31457280,
+      "step": 60
+    },
+    {
+      "epoch": 2.242491657397108,
+      "grad_norm": 0.416015625,
+      "learning_rate": 4.933523574614447e-06,
+      "loss": 2.6728,
+      "num_input_tokens_seen": 33030144,
+      "step": 63
+    },
+    {
+      "epoch": 2.349276974416018,
+      "grad_norm": 0.41015625,
+      "learning_rate": 3.6815306976265466e-06,
+      "loss": 2.6567,
+      "num_input_tokens_seen": 34603008,
+      "step": 66
+    },
+    {
+      "epoch": 2.456062291434928,
+      "grad_norm": 0.435546875,
+      "learning_rate": 2.590439511854144e-06,
+      "loss": 2.6862,
+      "num_input_tokens_seen": 36175872,
+      "step": 69
+    },
+    {
+      "epoch": 2.5628476084538376,
+      "grad_norm": 0.419921875,
+      "learning_rate": 1.6757608087630249e-06,
+      "loss": 2.6841,
+      "num_input_tokens_seen": 37748736,
+      "step": 72
+    },
+    {
+      "epoch": 2.6696329254727473,
+      "grad_norm": 0.40625,
+      "learning_rate": 9.504975259690835e-07,
+      "loss": 2.7272,
+      "num_input_tokens_seen": 39321600,
+      "step": 75
+    },
+    {
+      "epoch": 2.7764182424916575,
+      "grad_norm": 0.40625,
+      "learning_rate": 4.2495989939384916e-07,
+      "loss": 2.6643,
+      "num_input_tokens_seen": 40894464,
+      "step": 78
+    },
+    {
+      "epoch": 2.883203559510567,
+      "grad_norm": 0.408203125,
+      "learning_rate": 1.0661889447039886e-07,
+      "loss": 2.679,
+      "num_input_tokens_seen": 42467328,
+      "step": 81
+    },
+    {
+      "epoch": 2.9899888765294773,
+      "grad_norm": 0.412109375,
+      "learning_rate": 0.0,
+      "loss": 2.7116,
+      "num_input_tokens_seen": 44040192,
+      "step": 84
+    },
+    {
+      "epoch": 2.9899888765294773,
+      "eval_accuracy": 0.4297979192055684,
+      "eval_loss": 2.810983180999756,
+      "eval_runtime": 19.4794,
+      "eval_samples_per_second": 1.489,
+      "eval_steps_per_second": 1.489,
+      "num_input_tokens_seen": 44040192,
+      "step": 84
+    },
+    {
+      "epoch": 2.9899888765294773,
+      "num_input_tokens_seen": 44040192,
+      "step": 84,
+      "total_flos": 3.462459117703004e+17,
+      "train_loss": 2.7097946518943425,
+      "train_runtime": 6795.1554,
+      "train_samples_per_second": 0.397,
+      "train_steps_per_second": 0.012
+    }
+  ],
+  "logging_steps": 3,
+  "max_steps": 84,
+  "num_input_tokens_seen": 44040192,
+  "num_train_epochs": 3,
+  "save_steps": 50,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.462459117703004e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79753a25dbb8b2565e4d20ffdac00704b8318b52bea001e3e03550986f4e7bf0
+size 5240