Training in progress, epoch 1, checkpoint

checkpoint-2178/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-2178/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c766b5771d97b04eb191ccc7f88dba176ee104085f4dd36f59ec6129992f13d4
+size 143269386

checkpoint-2178/added_tokens.json

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

checkpoint-2178/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:18ae09f2a38f5b96ec4f2547feb27f5dc1fbd839fd78a0437600ffa6b2e3b161
-size 224537202
+oid sha256:01b86bec3bb9f176500ef2852dbda5d2be1b648c1be8b42fd07253229990e9fc
+size 286585234

checkpoint-2178/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5de6221fc07c1f6ddaf3c06156b6ded1ce97cf86a9903e6f9331d3a5369a5cc1
+oid sha256:b8ea0905fc13f0238ceb0842627ea7e9973738630616cb989b3d33771eaa864a
 size 15024

checkpoint-2178/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1077d721279e1085351a00f42116b5881bd4b1cd6bf9c7bc66bf77cb275a39d6
+oid sha256:c7e04982c20db1329f779195a444a3c7e439d4df1600843dfc773f785e8a6763
 size 15024

checkpoint-2178/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:968755856675429d912aae13ed83b470823e79849bc1165a0b9cd668c0d43fff
+oid sha256:a657ba51955b0761a65adc58d6700d4e9c74bf666222f674340466270b44d76c
 size 15024

checkpoint-2178/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:83bbb747a69b505f4f663785750dcf68996947076ec95bd2eb10ff136db94b7b
+oid sha256:9c579817b3fbd7915d0047259f2ef1d0a5457c792323e6b2478448c327d4bd84
 size 15024

checkpoint-2178/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-2178/tokenizer.model

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

checkpoint-2178/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-2178/trainer_state.json

@@ -9,202 +9,167 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.09180628873077806,
-      "grad_norm": 0.5307241678237915,
+      "epoch": 0.09,
       "learning_rate": 3e-05,
-      "loss": 2.0176,
+      "loss": 1.5477,
       "step": 100
     },
     {
-      "epoch": 0.18361257746155613,
-      "grad_norm": 0.6362708210945129,
+      "epoch": 0.18,
       "learning_rate": 3e-05,
-      "loss": 1.7953,
+      "loss": 1.3248,
       "step": 200
     },
     {
-      "epoch": 0.2754188661923342,
-      "grad_norm": 0.3823029696941376,
+      "epoch": 0.28,
       "learning_rate": 3e-05,
-      "loss": 1.7849,
+      "loss": 1.2394,
       "step": 300
     },
     {
-      "epoch": 0.36722515492311225,
-      "grad_norm": 0.43036168813705444,
+      "epoch": 0.37,
       "learning_rate": 3e-05,
-      "loss": 1.7818,
+      "loss": 1.2549,
       "step": 400
     },
     {
-      "epoch": 0.4590314436538903,
-      "grad_norm": 0.5922847390174866,
+      "epoch": 0.46,
       "learning_rate": 3e-05,
-      "loss": 1.7649,
+      "loss": 1.2163,
       "step": 500
     },
     {
-      "epoch": 0.5508377323846684,
-      "grad_norm": 0.40470582246780396,
+      "epoch": 0.55,
       "learning_rate": 3e-05,
-      "loss": 1.7902,
+      "loss": 1.2202,
       "step": 600
     },
     {
-      "epoch": 0.6426440211154464,
-      "grad_norm": 0.49739205837249756,
+      "epoch": 0.64,
       "learning_rate": 3e-05,
-      "loss": 1.7606,
+      "loss": 1.2099,
       "step": 700
     },
     {
-      "epoch": 0.7344503098462245,
-      "grad_norm": 0.43984296917915344,
+      "epoch": 0.73,
       "learning_rate": 3e-05,
-      "loss": 1.7746,
+      "loss": 1.2213,
       "step": 800
     },
     {
-      "epoch": 0.8262565985770025,
-      "grad_norm": 0.4566909968852997,
+      "epoch": 0.83,
       "learning_rate": 3e-05,
-      "loss": 1.7586,
+      "loss": 1.2203,
       "step": 900
     },
     {
-      "epoch": 0.9180628873077806,
-      "grad_norm": 0.4243898093700409,
+      "epoch": 0.92,
       "learning_rate": 3e-05,
-      "loss": 1.7775,
+      "loss": 1.2166,
       "step": 1000
     },
     {
-      "epoch": 0.999770484278173,
-      "eval_accuracy": 0.5157460317460317,
-      "eval_loss": 2.3358356952667236,
-      "eval_runtime": 5.2798,
-      "eval_samples_per_second": 94.7,
-      "eval_steps_per_second": 11.932,
+      "epoch": 1.0,
+      "eval_accuracy": 0.5925822784810126,
+      "eval_loss": 1.858124017715454,
+      "eval_runtime": 4.7843,
+      "eval_samples_per_second": 104.509,
+      "eval_steps_per_second": 13.168,
       "step": 1089
     },
     {
-      "epoch": 0.999770484278173,
-      "eval_exact_match": 19.0,
-      "eval_f1": 26.812075147075163,
+      "epoch": 1.0,
+      "eval_exact_match": 25.0,
+      "eval_f1": 33.90013632566266,
       "step": 1089
     },
     {
-      "epoch": 1.0098691760385587,
-      "grad_norm": 0.47265738248825073,
+      "epoch": 1.01,
       "learning_rate": 3e-05,
-      "loss": 1.7676,
+      "loss": 1.2141,
       "step": 1100
     },
     {
-      "epoch": 1.1016754647693368,
-      "grad_norm": 0.6015958189964294,
+      "epoch": 1.1,
       "learning_rate": 3e-05,
-      "loss": 1.7397,
+      "loss": 1.2178,
       "step": 1200
     },
     {
-      "epoch": 1.1934817535001148,
-      "grad_norm": 0.5499740242958069,
+      "epoch": 1.19,
       "learning_rate": 3e-05,
-      "loss": 1.7254,
+      "loss": 1.1986,
       "step": 1300
     },
     {
-      "epoch": 1.2852880422308928,
-      "grad_norm": 0.5272372961044312,
+      "epoch": 1.29,
       "learning_rate": 3e-05,
-      "loss": 1.73,
+      "loss": 1.1903,
       "step": 1400
     },
     {
-      "epoch": 1.377094330961671,
-      "grad_norm": 0.5437406897544861,
+      "epoch": 1.38,
       "learning_rate": 3e-05,
-      "loss": 1.7292,
+      "loss": 1.1925,
       "step": 1500
     },
     {
-      "epoch": 1.468900619692449,
-      "grad_norm": 0.5464481115341187,
+      "epoch": 1.47,
       "learning_rate": 3e-05,
-      "loss": 1.728,
+      "loss": 1.1954,
       "step": 1600
     },
     {
-      "epoch": 1.560706908423227,
-      "grad_norm": 0.6755242347717285,
+      "epoch": 1.56,
       "learning_rate": 3e-05,
-      "loss": 1.7407,
+      "loss": 1.1859,
       "step": 1700
     },
     {
-      "epoch": 1.652513197154005,
-      "grad_norm": 0.5893722176551819,
+      "epoch": 1.65,
       "learning_rate": 3e-05,
-      "loss": 1.7244,
+      "loss": 1.1979,
       "step": 1800
     },
     {
-      "epoch": 1.744319485884783,
-      "grad_norm": 0.5930923223495483,
+      "epoch": 1.74,
       "learning_rate": 3e-05,
-      "loss": 1.7225,
+      "loss": 1.201,
       "step": 1900
     },
     {
-      "epoch": 1.836125774615561,
-      "grad_norm": 0.5917518734931946,
+      "epoch": 1.84,
       "learning_rate": 3e-05,
-      "loss": 1.7189,
+      "loss": 1.1871,
       "step": 2000
     },
     {
-      "epoch": 1.9279320633463393,
-      "grad_norm": 0.6680920124053955,
+      "epoch": 1.93,
       "learning_rate": 3e-05,
-      "loss": 1.7313,
+      "loss": 1.1949,
       "step": 2100
     },
     {
-      "epoch": 1.999540968556346,
-      "eval_accuracy": 0.5178730158730158,
-      "eval_loss": 2.3068432807922363,
-      "eval_runtime": 5.4629,
-      "eval_samples_per_second": 91.527,
-      "eval_steps_per_second": 11.532,
+      "epoch": 2.0,
+      "eval_accuracy": 0.5960506329113924,
+      "eval_loss": 1.834545373916626,
+      "eval_runtime": 4.5511,
+      "eval_samples_per_second": 109.863,
+      "eval_steps_per_second": 13.843,
       "step": 2178
     },
     {
-      "epoch": 1.999540968556346,
-      "eval_exact_match": 17.8,
-      "eval_f1": 26.628214855320138,
+      "epoch": 2.0,
+      "eval_exact_match": 25.2,
+      "eval_f1": 34.41373827926461,
       "step": 2178
     }
   ],
   "logging_steps": 100,
   "max_steps": 54450,
-  "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": 1.5030257728670925e+17,
-  "train_batch_size": 1,
+  "total_flos": 4.1019316712872346e+17,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-2178/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5377ce2541570511bdbb29b699438bab217d7b0b49a543f6293cd608457a8a9d
-size 5304
+oid sha256:990ba0ec6800794569e1f142d0a98c1c27c344ee04644eb9619680965b92e617
+size 4728