Training in progress, step 4

README.md

@@ -27,7 +27,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/neko-llm/huggingface/runs/2w2mcwgz) 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/neko-llm/huggingface/runs/g4f13frh) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -25,10 +25,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "k_proj",
+    "q_proj",
     "o_proj",
-    "q_proj"
+    "k_proj",
+    "v_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a09d0d8b0c15f7e01381870495b3a9778b48adc01afb3ad86d661f4ac29d80b
+oid sha256:9b79c3a023aa6423ffdfa305a7a5615e42f3c90900c51afac55f44ad03f979eb
 size 101748240

all_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 2918992448585728.0,
+    "train_loss": 0.5622602303822836,
+    "train_runtime": 2418.6915,
+    "train_samples": 22441,
+    "train_samples_per_second": 0.238,
+    "train_steps_per_second": 0.001
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 2918992448585728.0,
+    "train_loss": 0.5622602303822836,
+    "train_runtime": 2418.6915,
+    "train_samples": 22441,
+    "train_samples_per_second": 0.238,
+    "train_steps_per_second": 0.001
+}

trainer_state.json

@@ -0,0 +1,67 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.13333333333333333,
+  "eval_steps": 500,
+  "global_step": 3,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.044444444444444446,
+      "grad_norm": 0.0809381034602598,
+      "learning_rate": 0.0,
+      "loss": 0.57,
+      "num_tokens": 6245325.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.08888888888888889,
+      "grad_norm": 0.07731609706200754,
+      "learning_rate": 1e-05,
+      "loss": 0.5777,
+      "num_tokens": 12529292.0,
+      "step": 2
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.07887771101388846,
+      "learning_rate": 5.500000000000001e-06,
+      "loss": 0.539,
+      "num_tokens": 18818029.0,
+      "step": 3
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "step": 3,
+      "total_flos": 2918992448585728.0,
+      "train_loss": 0.5622602303822836,
+      "train_runtime": 2418.6915,
+      "train_samples_per_second": 0.238,
+      "train_steps_per_second": 0.001
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 3,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 1,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2918992448585728.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0bd3ebf56d1b4392f28b225fcfd7080228386abb7948da578dd28bcc18fa25eb
+oid sha256:7e813c56ce63ac810e4a68cbc4af9209e456b93768e6c62d1e4318efd934c211
 size 7480