Training in progress, epoch 10, checkpoint

checkpoint-5290/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-5290/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e85ade6c7ffd589a6d69cdef902199fa87c0d5c5fb9d367218e3af0711ab72b9
+size 143269386

checkpoint-5290/added_tokens.json

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

checkpoint-5290/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d6e1e937dd122ad2127f64cb8ce9b0473347ae3002247fdd67212d1019f82b2d
-size 224537202
+oid sha256:200fea16632c265b293439d243e3f43f36a4b370c9d6b4d68305d5aec4ece7b6
+size 286585234

checkpoint-5290/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:603d6b52cc193d5c2a2efcfb5c13d5e227f74f70d83fac46c5387666aff41a2e
+oid sha256:873c515650bdd334a963554d5f96a1e24e3d28b954de188a8b43da4af411a7b2
 size 15024

checkpoint-5290/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:676bbd3244acb5b15b6460d0f9abe2bd17ec11c7bfdfdb164f2abb0ce6512391
+oid sha256:0b8d36fbb1c757dd8107f4cd3b0515d841f6a2df24fe74960107c84bb04ac742
 size 15024

checkpoint-5290/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:450ec0b3917ce9b1606aa464f8efcef060bce6a2e9eb4039f70e95ceb6d63412
+oid sha256:340e992257ccbfdc1a87ed1311619373c87fec52261387d5e4f105a7b9485a13
 size 15024

checkpoint-5290/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a5cdd4404d6cf4b7a493e9082bea38642a74fb4464d488bcf75015a379e34c6
+oid sha256:3a01099bbdb243f08272292c05a9e46560f99d0034f282a53c25e3bc4c4e050f
 size 15024

checkpoint-5290/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-5290/tokenizer.model

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

checkpoint-5290/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-5290/trainer_state.json

@@ -9,155 +9,140 @@
   "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
     },
     {
-      "epoch": 1.1342155009451795,
-      "grad_norm": 0.5318541526794434,
+      "epoch": 1.13,
       "learning_rate": 0.0005,
-      "loss": 1.6622,
+      "loss": 1.0305,
       "step": 600
     },
     {
-      "epoch": 1.3232514177693762,
-      "grad_norm": 235.6529998779297,
+      "epoch": 1.32,
       "learning_rate": 0.0005,
-      "loss": 1.5596,
+      "loss": 0.9199,
       "step": 700
     },
     {
-      "epoch": 1.5122873345935728,
-      "grad_norm": 0.5057249069213867,
+      "epoch": 1.51,
       "learning_rate": 0.0005,
-      "loss": 1.5296,
+      "loss": 0.9086,
       "step": 800
     },
     {
-      "epoch": 1.7013232514177694,
-      "grad_norm": 0.5339187383651733,
+      "epoch": 1.7,
       "learning_rate": 0.0005,
-      "loss": 1.5036,
+      "loss": 2.9245,
       "step": 900
     },
     {
-      "epoch": 1.8903591682419658,
-      "grad_norm": 0.5649825930595398,
+      "epoch": 1.89,
       "learning_rate": 0.0005,
-      "loss": 1.4884,
+      "loss": 4.9199,
       "step": 1000
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6643587443946188,
-      "eval_loss": 1.2189537286758423,
-      "eval_runtime": 6.192,
-      "eval_samples_per_second": 80.749,
-      "eval_steps_per_second": 10.174,
+      "eval_accuracy": 0.22922352941176471,
+      "eval_loss": 7.1224870681762695,
+      "eval_runtime": 7.7159,
+      "eval_samples_per_second": 64.801,
+      "eval_steps_per_second": 8.165,
       "step": 1058
     },
     {
       "epoch": 2.0,
-      "eval_exact_match": 11.2,
-      "eval_f1": 18.31238095238096,
-      "eval_qa_bleu": 10.766593815425745,
-      "eval_qa_exact_match": 0.096,
-      "eval_recite_bleu": 13.257993783037875,
-      "eval_recite_exact_match": 0.002,
+      "eval_exact_match": 0.0,
+      "eval_f1": 0.0,
+      "eval_qa_bleu": 0.0,
+      "eval_qa_exact_match": 0.0,
+      "eval_recite_bleu": 0.0,
+      "eval_recite_exact_match": 0.0,
       "step": 1058
     },
     {
-      "epoch": 2.0793950850661624,
-      "grad_norm": 0.6042512655258179,
+      "epoch": 2.08,
       "learning_rate": 0.0005,
-      "loss": 1.2966,
+      "loss": 7.4349,
       "step": 1100
     },
     {
-      "epoch": 2.268431001890359,
-      "grad_norm": 965.2459106445312,
+      "epoch": 2.27,
       "learning_rate": 0.0005,
-      "loss": 8.8153,
+      "loss": 8.0921,
       "step": 1200
     },
     {
-      "epoch": 2.4574669187145557,
-      "grad_norm": 3.0128896236419678,
+      "epoch": 2.46,
       "learning_rate": 0.0005,
-      "loss": 9.1241,
+      "loss": 7.889,
       "step": 1300
     },
     {
-      "epoch": 2.6465028355387523,
-      "grad_norm": 108062179328.0,
+      "epoch": 2.65,
       "learning_rate": 0.0005,
-      "loss": 10.5947,
+      "loss": 7.2559,
       "step": 1400
     },
     {
-      "epoch": 2.835538752362949,
-      "grad_norm": 5002999889920.0,
+      "epoch": 2.84,
       "learning_rate": 0.0005,
-      "loss": 12.5972,
+      "loss": 7.3563,
       "step": 1500
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.21054708520179372,
-      "eval_loss": 19.724363327026367,
-      "eval_runtime": 6.5285,
-      "eval_samples_per_second": 76.587,
-      "eval_steps_per_second": 9.65,
+      "eval_accuracy": 0.24235294117647058,
+      "eval_loss": 7.308440685272217,
+      "eval_runtime": 7.7087,
+      "eval_samples_per_second": 64.862,
+      "eval_steps_per_second": 8.173,
       "step": 1587
     },
     {
@@ -171,54 +156,48 @@
       "step": 1587
     },
     {
-      "epoch": 3.0245746691871456,
-      "grad_norm": 62.19309616088867,
+      "epoch": 3.02,
       "learning_rate": 0.0005,
-      "loss": 13.3192,
+      "loss": 7.2721,
       "step": 1600
     },
     {
-      "epoch": 3.213610586011342,
-      "grad_norm": 4.109750747680664,
+      "epoch": 3.21,
       "learning_rate": 0.0005,
-      "loss": 8.4804,
+      "loss": 7.2743,
       "step": 1700
     },
     {
-      "epoch": 3.402646502835539,
-      "grad_norm": 7.620689392089844,
+      "epoch": 3.4,
       "learning_rate": 0.0005,
-      "loss": 8.347,
+      "loss": 7.3151,
       "step": 1800
     },
     {
-      "epoch": 3.5916824196597354,
-      "grad_norm": 33.092830657958984,
+      "epoch": 3.59,
       "learning_rate": 0.0005,
-      "loss": 7.4667,
+      "loss": 7.3366,
       "step": 1900
     },
     {
-      "epoch": 3.780718336483932,
-      "grad_norm": 204828.203125,
+      "epoch": 3.78,
       "learning_rate": 0.0005,
-      "loss": 7.885,
+      "loss": 7.3294,
       "step": 2000
     },
     {
-      "epoch": 3.9697542533081287,
-      "grad_norm": 229958729728.0,
+      "epoch": 3.97,
       "learning_rate": 0.0005,
-      "loss": 8.4174,
+      "loss": 7.3158,
       "step": 2100
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.22234977578475337,
-      "eval_loss": 9.6956148147583,
-      "eval_runtime": 5.4245,
-      "eval_samples_per_second": 92.174,
-      "eval_steps_per_second": 11.614,
+      "eval_accuracy": 0.23965490196078432,
+      "eval_loss": 7.242833614349365,
+      "eval_runtime": 6.6551,
+      "eval_samples_per_second": 75.13,
+      "eval_steps_per_second": 9.466,
       "step": 2116
     },
     {
@@ -232,47 +211,42 @@
       "step": 2116
     },
     {
-      "epoch": 4.158790170132325,
-      "grad_norm": 17.098289489746094,
+      "epoch": 4.16,
       "learning_rate": 0.0005,
-      "loss": 10.4359,
+      "loss": 7.3307,
       "step": 2200
     },
     {
-      "epoch": 4.3478260869565215,
-      "grad_norm": 1069.9039306640625,
+      "epoch": 4.35,
       "learning_rate": 0.0005,
-      "loss": 9.5369,
+      "loss": 7.3307,
       "step": 2300
     },
     {
-      "epoch": 4.536862003780718,
-      "grad_norm": 7.694222927093506,
+      "epoch": 4.54,
       "learning_rate": 0.0005,
-      "loss": 7.7028,
+      "loss": 7.3197,
       "step": 2400
     },
     {
-      "epoch": 4.725897920604915,
-      "grad_norm": Infinity,
+      "epoch": 4.73,
       "learning_rate": 0.0005,
-      "loss": 8.4852,
+      "loss": 7.3238,
       "step": 2500
     },
     {
-      "epoch": 4.914933837429111,
-      "grad_norm": 396939770200064.0,
+      "epoch": 4.91,
       "learning_rate": 0.0005,
-      "loss": 9.5996,
+      "loss": 7.3017,
       "step": 2600
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.2166188340807175,
-      "eval_loss": 10.412571907043457,
-      "eval_runtime": 5.8473,
-      "eval_samples_per_second": 85.51,
-      "eval_steps_per_second": 10.774,
+      "eval_accuracy": 0.24474509803921568,
+      "eval_loss": 7.229503631591797,
+      "eval_runtime": 7.0755,
+      "eval_samples_per_second": 70.667,
+      "eval_steps_per_second": 8.904,
       "step": 2645
     },
     {
@@ -286,47 +260,42 @@
       "step": 2645
     },
     {
-      "epoch": 5.103969754253308,
-      "grad_norm": 3.166937635134374e+18,
+      "epoch": 5.1,
       "learning_rate": 0.0005,
-      "loss": 10.3832,
+      "loss": 7.2926,
       "step": 2700
     },
     {
-      "epoch": 5.293005671077505,
-      "grad_norm": 1.8363503072027607e+18,
+      "epoch": 5.29,
       "learning_rate": 0.0005,
-      "loss": 9.7717,
+      "loss": 7.3006,
       "step": 2800
     },
     {
-      "epoch": 5.482041587901701,
-      "grad_norm": 113262112800768.0,
+      "epoch": 5.48,
       "learning_rate": 0.0005,
-      "loss": 8.3066,
+      "loss": 7.3051,
       "step": 2900
     },
     {
-      "epoch": 5.671077504725898,
-      "grad_norm": 1283434880.0,
+      "epoch": 5.67,
       "learning_rate": 0.0005,
-      "loss": 8.6003,
+      "loss": 7.2973,
       "step": 3000
     },
     {
-      "epoch": 5.8601134215500945,
-      "grad_norm": 1221250.5,
+      "epoch": 5.86,
       "learning_rate": 0.0005,
-      "loss": 9.1302,
+      "loss": 7.281,
       "step": 3100
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.21948878923766815,
-      "eval_loss": 9.434032440185547,
-      "eval_runtime": 6.1208,
-      "eval_samples_per_second": 81.689,
-      "eval_steps_per_second": 10.293,
+      "eval_accuracy": 0.24474509803921568,
+      "eval_loss": 7.229503631591797,
+      "eval_runtime": 7.1043,
+      "eval_samples_per_second": 70.38,
+      "eval_steps_per_second": 8.868,
       "step": 3174
     },
     {
@@ -340,54 +309,48 @@
       "step": 3174
     },
     {
-      "epoch": 6.049149338374291,
-      "grad_norm": 2019.499267578125,
+      "epoch": 6.05,
       "learning_rate": 0.0005,
-      "loss": 9.3999,
+      "loss": 7.3017,
       "step": 3200
     },
     {
-      "epoch": 6.238185255198488,
-      "grad_norm": 164.9034881591797,
+      "epoch": 6.24,
       "learning_rate": 0.0005,
-      "loss": 8.1768,
+      "loss": 7.297,
       "step": 3300
     },
     {
-      "epoch": 6.427221172022684,
-      "grad_norm": 434.93157958984375,
+      "epoch": 6.43,
       "learning_rate": 0.0005,
-      "loss": 7.6329,
+      "loss": 7.2877,
       "step": 3400
     },
     {
-      "epoch": 6.616257088846881,
-      "grad_norm": 1324.7332763671875,
+      "epoch": 6.62,
       "learning_rate": 0.0005,
-      "loss": 7.7765,
+      "loss": 7.296,
       "step": 3500
     },
     {
-      "epoch": 6.805293005671078,
-      "grad_norm": 52531002605568.0,
+      "epoch": 6.81,
       "learning_rate": 0.0005,
-      "loss": 8.006,
+      "loss": 7.2993,
       "step": 3600
     },
     {
-      "epoch": 6.994328922495274,
-      "grad_norm": Infinity,
+      "epoch": 6.99,
       "learning_rate": 0.0005,
-      "loss": 15.3926,
+      "loss": 7.3056,
       "step": 3700
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.23054708520179373,
-      "eval_loss": 15.052428245544434,
-      "eval_runtime": 5.4671,
-      "eval_samples_per_second": 91.457,
-      "eval_steps_per_second": 11.524,
+      "eval_accuracy": 0.24474509803921568,
+      "eval_loss": 7.229503631591797,
+      "eval_runtime": 7.6931,
+      "eval_samples_per_second": 64.993,
+      "eval_steps_per_second": 8.189,
       "step": 3703
     },
     {
@@ -401,47 +364,42 @@
       "step": 3703
     },
     {
-      "epoch": 7.183364839319471,
-      "grad_norm": Infinity,
+      "epoch": 7.18,
       "learning_rate": 0.0005,
-      "loss": 17.6263,
+      "loss": 7.299,
       "step": 3800
     },
     {
-      "epoch": 7.3724007561436675,
-      "grad_norm": Infinity,
+      "epoch": 7.37,
       "learning_rate": 0.0005,
-      "loss": 17.7669,
+      "loss": 7.2954,
       "step": 3900
     },
     {
-      "epoch": 7.561436672967864,
-      "grad_norm": Infinity,
+      "epoch": 7.56,
       "learning_rate": 0.0005,
-      "loss": 17.7462,
+      "loss": 7.3002,
       "step": 4000
     },
     {
-      "epoch": 7.750472589792061,
-      "grad_norm": Infinity,
+      "epoch": 7.75,
       "learning_rate": 0.0005,
-      "loss": 17.6186,
+      "loss": 7.3011,
       "step": 4100
     },
     {
-      "epoch": 7.939508506616257,
-      "grad_norm": NaN,
+      "epoch": 7.94,
       "learning_rate": 0.0005,
-      "loss": 1574.3041,
+      "loss": 7.2943,
       "step": 4200
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.21060089686098654,
-      "eval_loss": NaN,
-      "eval_runtime": 5.4376,
-      "eval_samples_per_second": 91.952,
-      "eval_steps_per_second": 11.586,
+      "eval_accuracy": 0.24474509803921568,
+      "eval_loss": 7.229503631591797,
+      "eval_runtime": 6.6465,
+      "eval_samples_per_second": 75.228,
+      "eval_steps_per_second": 9.479,
       "step": 4232
     },
     {
@@ -455,47 +413,42 @@
       "step": 4232
     },
     {
-      "epoch": 8.128544423440454,
-      "grad_norm": NaN,
+      "epoch": 8.13,
       "learning_rate": 0.0005,
-      "loss": 0.0,
+      "loss": 7.3001,
       "step": 4300
     },
     {
-      "epoch": 8.31758034026465,
-      "grad_norm": NaN,
+      "epoch": 8.32,
       "learning_rate": 0.0005,
-      "loss": 0.0,
+      "loss": 7.2922,
       "step": 4400
     },
     {
-      "epoch": 8.506616257088847,
-      "grad_norm": NaN,
+      "epoch": 8.51,
       "learning_rate": 0.0005,
-      "loss": 0.0,
+      "loss": 7.2969,
       "step": 4500
     },
     {
-      "epoch": 8.695652173913043,
-      "grad_norm": NaN,
+      "epoch": 8.7,
       "learning_rate": 0.0005,
-      "loss": 0.0,
+      "loss": 7.2879,
       "step": 4600
     },
     {
-      "epoch": 8.88468809073724,
-      "grad_norm": NaN,
+      "epoch": 8.88,
       "learning_rate": 0.0005,
-      "loss": 0.0,
+      "loss": 7.3011,
       "step": 4700
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.21060089686098654,
-      "eval_loss": NaN,
-      "eval_runtime": 6.1031,
-      "eval_samples_per_second": 81.925,
-      "eval_steps_per_second": 10.323,
+      "eval_accuracy": 0.24474509803921568,
+      "eval_loss": 7.229503631591797,
+      "eval_runtime": 7.77,
+      "eval_samples_per_second": 64.35,
+      "eval_steps_per_second": 8.108,
       "step": 4761
     },
     {
@@ -509,47 +462,42 @@
       "step": 4761
     },
     {
-      "epoch": 9.073724007561436,
-      "grad_norm": NaN,
+      "epoch": 9.07,
       "learning_rate": 0.0005,
-      "loss": 0.0,
+      "loss": 7.3066,
       "step": 4800
     },
     {
-      "epoch": 9.262759924385634,
-      "grad_norm": NaN,
+      "epoch": 9.26,
       "learning_rate": 0.0005,
-      "loss": 0.0,
+      "loss": 7.2949,
       "step": 4900
     },
     {
-      "epoch": 9.45179584120983,
-      "grad_norm": NaN,
+      "epoch": 9.45,
       "learning_rate": 0.0005,
-      "loss": 0.0,
+      "loss": 7.2955,
       "step": 5000
     },
     {
-      "epoch": 9.640831758034027,
-      "grad_norm": NaN,
+      "epoch": 9.64,
       "learning_rate": 0.0005,
-      "loss": 0.0,
+      "loss": 7.3061,
       "step": 5100
     },
     {
-      "epoch": 9.829867674858223,
-      "grad_norm": NaN,
+      "epoch": 9.83,
       "learning_rate": 0.0005,
-      "loss": 0.0,
+      "loss": 7.2948,
       "step": 5200
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.21060089686098654,
-      "eval_loss": NaN,
-      "eval_runtime": 5.5143,
-      "eval_samples_per_second": 90.673,
-      "eval_steps_per_second": 11.425,
+      "eval_accuracy": 0.24474509803921568,
+      "eval_loss": 7.229503631591797,
+      "eval_runtime": 7.0663,
+      "eval_samples_per_second": 70.758,
+      "eval_steps_per_second": 8.916,
       "step": 5290
     },
     {
@@ -565,23 +513,9 @@
   ],
   "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": 4.821221566589174e+17,
-  "train_batch_size": 1,
+  "total_flos": 1.1216445557880914e+18,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-5290/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