Training in progress, epoch 2, checkpoint

checkpoint-375/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-375/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bea5be323103b911e888bb13f2500999a6bb96b4d9af32d9ae47e50b53289479
+size 143269386

checkpoint-375/added_tokens.json

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

checkpoint-375/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5878dc16b1a31c628d86ef480c90810eb669c72ad5ea8472e8e1a3fdd332f9a
-size 224537202
+oid sha256:ff951a2ad69ce8366421f6ab12fce1f59ea33181dd61a8a9987df4a353b84c9c
+size 286585234

checkpoint-375/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d698b6cca1dda1cc04538825d2269c05e6f76453e50e98adf71d44d36d0f9b4
+oid sha256:88853c02af627041443826b1117e27ac3d5514d6c00f91fc179a028215d747f0
 size 15024

checkpoint-375/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae391cc53ee4121fdad983e1a28a29c5dfbd29b463abaf9ab787242819227522
+oid sha256:b84929911a75a960a78ade0e35565e09aa2b00f96968cc5e881b061e4c07019d
 size 15024

checkpoint-375/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48c302a008780224918f0b45ed64b1e93732ba2ae5bed1e49079ebd7be5ac30b
+oid sha256:5ae575534afd087050efebb7921a11b74db4fa5b118752e868129b6989edd87f
 size 15024

checkpoint-375/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:287f42204497d8f9fb9a83ab4b44a73237ac9e378e18a78f55567a10d0c44cbc
+oid sha256:9e5ed8967bfa708d2ea95b00e16cc466f261842b6e13f7429129596cdca090c0
 size 15024

checkpoint-375/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-375/tokenizer.model

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

checkpoint-375/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-375/trainer_state.json

@@ -9,76 +9,59 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.5333333333333333,
-      "grad_norm": 0.64143306016922,
+      "epoch": 0.53,
       "learning_rate": 0.0001,
-      "loss": 1.7779,
+      "loss": 1.598,
       "step": 100
     },
     {
-      "epoch": 0.9973333333333333,
-      "eval_accuracy": 0.5734871794871795,
-      "eval_loss": 1.6191272735595703,
-      "eval_runtime": 5.393,
-      "eval_samples_per_second": 92.713,
-      "eval_steps_per_second": 11.682,
+      "epoch": 1.0,
+      "eval_accuracy": 0.6146666666666667,
+      "eval_loss": 1.2691738605499268,
+      "eval_runtime": 4.3939,
+      "eval_samples_per_second": 113.793,
+      "eval_steps_per_second": 14.338,
       "step": 187
     },
     {
-      "epoch": 0.9973333333333333,
-      "eval_exact_match": 18.0,
-      "eval_f1": 28.280476190476193,
+      "epoch": 1.0,
+      "eval_exact_match": 24.8,
+      "eval_f1": 38.10857142857143,
       "step": 187
     },
     {
-      "epoch": 1.0666666666666667,
-      "grad_norm": 0.7118939161300659,
+      "epoch": 1.07,
       "learning_rate": 0.0001,
-      "loss": 1.6349,
+      "loss": 1.2894,
       "step": 200
     },
     {
       "epoch": 1.6,
-      "grad_norm": 1.0694799423217773,
       "learning_rate": 0.0001,
-      "loss": 1.4887,
+      "loss": 1.1923,
       "step": 300
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5735897435897436,
-      "eval_loss": 1.6172032356262207,
-      "eval_runtime": 5.0425,
-      "eval_samples_per_second": 99.158,
-      "eval_steps_per_second": 12.494,
+      "eval_accuracy": 0.6175897435897436,
+      "eval_loss": 1.2732893228530884,
+      "eval_runtime": 4.5482,
+      "eval_samples_per_second": 109.934,
+      "eval_steps_per_second": 13.852,
       "step": 375
     },
     {
       "epoch": 2.0,
-      "eval_exact_match": 19.8,
-      "eval_f1": 28.77095238095238,
+      "eval_exact_match": 25.6,
+      "eval_f1": 39.04746031746033,
       "step": 375
     }
   ],
   "logging_steps": 100,
   "max_steps": 9350,
-  "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": 6864138500308992.0,
-  "train_batch_size": 2,
+  "total_flos": 1.3997758247600128e+16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-375/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d667d5e0e40fe951f9fe0d2cd7fb4682a608b09af78a113a245307211adedfbc
-size 5304
+oid sha256:dfdad39e2291c20f000713e6b820dbeb747acc6873c9eec22bd69c854755aa38
+size 4728