Training in progress, epoch 1, checkpoint

checkpoint-529/adapter_config.json

@@ -1,6 +1,6 @@
 {
   "auto_mapping": null,
-  "base_model_name_or_path": "Qwen/Qwen1.5-4B",
+  "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,

checkpoint-529/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ae384f706f9c48efd72301691f5d80739ed9a354e0fbde6eb942171e74a70c0
+size 143269386

checkpoint-529/added_tokens.json

@@ -1,5 +1,5 @@
 {
-  "<|endoftext|>": 151643,
-  "<|im_end|>": 151645,
-  "<|im_start|>": 151644
+  "</s>": 2,
+  "<s>": 1,
+  "<unk>": 0
 }

checkpoint-529/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b7a98f33c59bb91004aeebe017a0b339ef2219e1fc8e0db0cbd68ed122942b6
-size 224537202
+oid sha256:f0b6fb5747f1215400915699206a34bda452b6f204455ab7bfd5e7cddeaaf83b
+size 286585234

checkpoint-529/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7a273358f9ce09aa62e77874a78a0741af575acbda1a439ac26180982bfc8df
+oid sha256:0f479d58d5361b7916de7a90abac6cef2ba4fe0b85e39e360c170436fc3872f0
 size 15024

checkpoint-529/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e888d51d55f94cd3341c18abe0387d9aa576daad70d38337424d3989a3f65d89
+oid sha256:c31fc4f1ecf863c47ac3057ffe8a8b58dbf4112e245a262cd597eaf1cd17c4dc
 size 15024

checkpoint-529/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c2bcefeff236c6f69049e47dc88b18c1356634ae6d0f3c98d19bc52599e8b4a
+oid sha256:5dbc360c1e758044ba7dcc42b700087cc5540d5952d683ade28ea9048a0c9449
 size 15024

checkpoint-529/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68af46d9ac10ac3769c613c9f73ae8ceab66ca499707dd4896a7f63af0ebaf62
+oid sha256:5e1118e1e96ffcab25a148734ee4c44e880d1f64127f25b7ba403c6e0ee64b6e
 size 15024

checkpoint-529/special_tokens_map.json

@@ -1,14 +1,6 @@
 {
-  "additional_special_tokens": [
-    "<|im_start|>",
-    "<|im_end|>"
-  ],
-  "eos_token": {
-    "content": "<|endoftext|>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  },
-  "pad_token": "<|endoftext|>"
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "</s>",
+  "unk_token": "<unk>"
 }

checkpoint-529/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

checkpoint-529/tokenizer_config.json

@@ -1,24 +1,23 @@
 {
-  "add_prefix_space": false,
   "added_tokens_decoder": {
-    "151643": {
-      "content": "<|endoftext|>",
+    "0": {
+      "content": "<unk>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
-    "151644": {
-      "content": "<|im_start|>",
+    "1": {
+      "content": "<s>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
-    "151645": {
-      "content": "<|im_end|>",
+    "2": {
+      "content": "</s>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -26,18 +25,16 @@
       "special": true
     }
   },
-  "additional_special_tokens": [
-    "<|im_start|>",
-    "<|im_end|>"
-  ],
-  "bos_token": null,
-  "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
   "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
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "left",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true
 }

checkpoint-529/trainer_state.json

@@ -9,75 +9,56 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.1890359168241966,
-      "grad_norm": 0.5107442140579224,
+      "epoch": 0.19,
       "learning_rate": 5e-05,
-      "loss": 2.0482,
+      "loss": 1.6355,
       "step": 100
     },
     {
-      "epoch": 0.3780718336483932,
-      "grad_norm": 0.5563424229621887,
+      "epoch": 0.38,
       "learning_rate": 5e-05,
-      "loss": 1.8835,
+      "loss": 1.4037,
       "step": 200
     },
     {
-      "epoch": 0.5671077504725898,
-      "grad_norm": 0.4735371172428131,
+      "epoch": 0.57,
       "learning_rate": 5e-05,
-      "loss": 1.8512,
+      "loss": 1.4183,
       "step": 300
     },
     {
-      "epoch": 0.7561436672967864,
-      "grad_norm": 0.4543495178222656,
+      "epoch": 0.76,
       "learning_rate": 5e-05,
-      "loss": 1.8489,
+      "loss": 1.3881,
       "step": 400
     },
     {
-      "epoch": 0.945179584120983,
-      "grad_norm": 0.5007112622261047,
+      "epoch": 0.95,
       "learning_rate": 5e-05,
-      "loss": 1.8439,
+      "loss": 1.3891,
       "step": 500
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5733333333333334,
-      "eval_loss": 1.6194567680358887,
-      "eval_runtime": 5.4154,
-      "eval_samples_per_second": 92.33,
-      "eval_steps_per_second": 11.634,
+      "eval_accuracy": 0.6137948717948718,
+      "eval_loss": 1.3015451431274414,
+      "eval_runtime": 4.1821,
+      "eval_samples_per_second": 119.557,
+      "eval_steps_per_second": 15.064,
       "step": 529
     },
     {
       "epoch": 1.0,
-      "eval_exact_match": 17.0,
-      "eval_f1": 26.880952380952387,
+      "eval_exact_match": 23.2,
+      "eval_f1": 35.17014759120022,
       "step": 529
     }
   ],
   "logging_steps": 100,
   "max_steps": 26450,
-  "num_input_tokens_seen": 0,
   "num_train_epochs": 50,
   "save_steps": 500,
-  "stateful_callbacks": {
-    "TrainerControl": {
-      "args": {
-        "should_epoch_stop": false,
-        "should_evaluate": false,
-        "should_log": false,
-        "should_save": true,
-        "should_training_stop": false
-      },
-      "attributes": {}
-    }
-  },
-  "total_flos": 3.258622154899456e+16,
-  "train_batch_size": 1,
+  "total_flos": 8.922877128186266e+16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-529/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f01d18f717251d8791982fc93ed5352f2d247add98e1102e2a52cc0a32128bb7
-size 5304
+oid sha256:391c04887a1ac777804ab56ff52563bcf9c4927713b353f37947aac9be091742
+size 4728