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:a89733684743699caebbf287bf33066ae05c094fc8b0f71a42a78182c6d82588
+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 sha256:59109826c6966f45b4bd254fc223707e9a488a95a06f39767e0f3924da4698ac
-size 224537202
+oid sha256:6f1d4442625ebbdac25f6a87d9220fe5ab74d2489c5235c32fa38da28c19b7fb
+size 286585234

checkpoint-529/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:78eae6532100bfa57e11db78d54753d933bd17fb49dd7d2436138521bff5039f
+oid sha256:a9e14651318e23d70dc44585f7458ec8369236cc3fa8ec21e4d5a2cbd2748e2d
 size 15024

checkpoint-529/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e6ded07878f46a76371af2a108c6667c90721d5df8b5c12104627264f83b4ce
+oid sha256:af6bd8f21e81a1d80426cf9e4fb8a331e2a1e547bf8c009328f3b683421d9804
 size 15024

checkpoint-529/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac152ca659c4d074b0f088a2f48c6e0d86bc029066af52944d1fc34f8bce3a73
+oid sha256:cb8692f391880ad2969d738cf6c2c4fb9b2d1e6793e5d787311f036067ceffc5
 size 15024

checkpoint-529/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b412ed3e09338e6b2ab9a96b9c632376c2c67f266e4c543cf8f4e14d3285765
+oid sha256:1c5271a2cc2e74a2ea00bc6d3c8455bc5749e86eadab5748955c539c22d1d5af
 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

@@ -9,79 +9,60 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.1890359168241966,
-      "grad_norm": 7.269421577453613,
+      "epoch": 0.19,
       "learning_rate": 0.0005,
-      "loss": 1.9195,
+      "loss": 1.3631,
       "step": 100
     },
     {
-      "epoch": 0.3780718336483932,
-      "grad_norm": 1.071650505065918,
+      "epoch": 0.38,
       "learning_rate": 0.0005,
-      "loss": 1.9396,
+      "loss": 1.294,
       "step": 200
     },
     {
-      "epoch": 0.5671077504725898,
-      "grad_norm": 0.5662499070167542,
+      "epoch": 0.57,
       "learning_rate": 0.0005,
-      "loss": 2.3641,
+      "loss": 1.302,
       "step": 300
     },
     {
-      "epoch": 0.7561436672967864,
-      "grad_norm": 0.47088223695755005,
+      "epoch": 0.76,
       "learning_rate": 0.0005,
-      "loss": 1.8969,
+      "loss": 1.2626,
       "step": 400
     },
     {
-      "epoch": 0.945179584120983,
-      "grad_norm": 0.3653932511806488,
+      "epoch": 0.95,
       "learning_rate": 0.0005,
-      "loss": 1.8531,
+      "loss": 1.2156,
       "step": 500
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6214529147982063,
-      "eval_loss": 1.5580899715423584,
-      "eval_runtime": 5.56,
-      "eval_samples_per_second": 89.928,
-      "eval_steps_per_second": 11.331,
+      "eval_accuracy": 0.6911294117647059,
+      "eval_loss": 1.012805700302124,
+      "eval_runtime": 7.0918,
+      "eval_samples_per_second": 70.504,
+      "eval_steps_per_second": 8.883,
       "step": 529
     },
     {
       "epoch": 1.0,
-      "eval_exact_match": 14.4,
-      "eval_f1": 20.983333333333345,
-      "eval_qa_bleu": 12.514681390986086,
-      "eval_qa_exact_match": 0.102,
-      "eval_recite_bleu": 12.61307158352263,
-      "eval_recite_exact_match": 0.0,
+      "eval_exact_match": 27.6,
+      "eval_f1": 35.68145743145743,
+      "eval_qa_bleu": 19.064111689202868,
+      "eval_qa_exact_match": 0.216,
+      "eval_recite_bleu": 17.898525975872754,
+      "eval_recite_exact_match": 0.006,
       "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": 4.821247382362522e+16,
-  "train_batch_size": 1,
+  "total_flos": 1.1216126490129203e+17,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-529/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1820a0b512c2fa8b3950f986e6f44420dacfa98792c1f255a5842682ec51f235
-size 5304
+oid sha256:b01f751e6b30f6919dad64453b51b9e04d3c36ac30d2f54e574df56cf37b82bc
+size 4728