test - Updated Model

adapter_config.json

@@ -1,7 +1,7 @@
 {
   "alpha_pattern": {},
   "auto_mapping": null,
-  "base_model_name_or_path": "unsloth/gemma-1.1-7b-it-bnb-4bit",
+  "base_model_name_or_path": "google/gemma-1.1-7b-it",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,
@@ -10,23 +10,23 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 16,
-  "lora_dropout": 0,
+  "lora_alpha": 32,
+  "lora_dropout": 0.0,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 16,
+  "r": 32,
   "rank_pattern": {},
-  "revision": "unsloth",
+  "revision": null,
   "target_modules": [
-    "gate_proj",
-    "down_proj",
-    "k_proj",
+    "v_proj",
     "o_proj",
+    "gate_proj",
+    "q_proj",
     "up_proj",
-    "v_proj",
-    "q_proj"
+    "k_proj",
+    "down_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9fb422ca9ce78def394728e4c71ac93f2818fffd892735a86fea01ae02245cf8
-size 200068512
+oid sha256:1436fec4bc48f80ece49239fa36dbf60e970248628bf31230321a5484df168ba
+size 200068904

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0da58d42d70e972774acd49a657d758c19543e002e304cb44a62f791de875959
+size 200871604

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:386fcc8cc1089aade9450d86fb239ea3483f455fd2d78d8378645feecfec9d69
+size 14244

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:802f82de1a07e7c596638d1365d402f6236ba5f0d2c0221ccaab7fc415f74306
+size 1064

tokenizer_config.json

@@ -1747,9 +1747,8 @@
   "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<eos>",
-  "model_max_length": 8192,
+  "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<pad>",
-  "padding_side": "right",
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,
   "tokenizer_class": "GemmaTokenizer",

trainer_state.json

@@ -0,0 +1,208 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 12500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "grad_norm": 2.125,
+      "learning_rate": 9.603841536614646e-05,
+      "loss": 1.8394,
+      "step": 500
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 2.046875,
+      "learning_rate": 9.203681472589036e-05,
+      "loss": 1.3706,
+      "step": 1000
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 1.8984375,
+      "learning_rate": 8.803521408563425e-05,
+      "loss": 1.3084,
+      "step": 1500
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 2.046875,
+      "learning_rate": 8.403361344537815e-05,
+      "loss": 1.2739,
+      "step": 2000
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 1.734375,
+      "learning_rate": 8.003201280512204e-05,
+      "loss": 1.2551,
+      "step": 2500
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 2.671875,
+      "learning_rate": 7.603041216486595e-05,
+      "loss": 1.2183,
+      "step": 3000
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 1.7421875,
+      "learning_rate": 7.202881152460985e-05,
+      "loss": 1.1962,
+      "step": 3500
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 1.8828125,
+      "learning_rate": 6.802721088435374e-05,
+      "loss": 1.1709,
+      "step": 4000
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 1.9765625,
+      "learning_rate": 6.402561024409764e-05,
+      "loss": 1.1534,
+      "step": 4500
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 2.078125,
+      "learning_rate": 6.002400960384153e-05,
+      "loss": 1.1418,
+      "step": 5000
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 2.765625,
+      "learning_rate": 5.6022408963585436e-05,
+      "loss": 1.1375,
+      "step": 5500
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 1.9453125,
+      "learning_rate": 5.202080832332934e-05,
+      "loss": 1.1271,
+      "step": 6000
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 1.9453125,
+      "learning_rate": 4.801920768307323e-05,
+      "loss": 1.0945,
+      "step": 6500
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 2.03125,
+      "learning_rate": 4.4017607042817125e-05,
+      "loss": 1.106,
+      "step": 7000
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 2.296875,
+      "learning_rate": 4.001600640256102e-05,
+      "loss": 1.0835,
+      "step": 7500
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 1.90625,
+      "learning_rate": 3.6014405762304925e-05,
+      "loss": 1.0687,
+      "step": 8000
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 2.15625,
+      "learning_rate": 3.201280512204882e-05,
+      "loss": 1.0637,
+      "step": 8500
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 1.8984375,
+      "learning_rate": 2.8011204481792718e-05,
+      "loss": 1.0477,
+      "step": 9000
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 2.546875,
+      "learning_rate": 2.4009603841536614e-05,
+      "loss": 1.0494,
+      "step": 9500
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 1.890625,
+      "learning_rate": 2.000800320128051e-05,
+      "loss": 1.0529,
+      "step": 10000
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 1.6484375,
+      "learning_rate": 1.600640256102441e-05,
+      "loss": 1.0219,
+      "step": 10500
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 1.8359375,
+      "learning_rate": 1.2004801920768307e-05,
+      "loss": 1.0338,
+      "step": 11000
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 2.125,
+      "learning_rate": 8.003201280512205e-06,
+      "loss": 1.0454,
+      "step": 11500
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 1.828125,
+      "learning_rate": 4.001600640256103e-06,
+      "loss": 1.0405,
+      "step": 12000
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.8359375,
+      "learning_rate": 0.0,
+      "loss": 1.0363,
+      "step": 12500
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 12500,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.596194976892191e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:203c6f7d5f7372475862264ad298c4e3c9d162eadf7ba0985dd1957886cec21a
+size 5112