Training in progress, step 1400, checkpoint

last-checkpoint/README.md

@@ -201,6 +201,25 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
 ## Training procedure
 
 
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
+
+### Framework versions
+
+
+- PEFT 0.6.0.dev0
+## Training procedure
+
+
 ### Framework versions
 
 

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:05000e36c627807d10f5b2a2a12fef5b7ae8bc4f815bc0f4baef1067e14f8a7e
+oid sha256:287cdc0db0836385fc8b23617ba8d112ec3a39080cdfde1f53ea06dc230519e9
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a255d88304c99a62eb2a041b92e091d0389471f9d3d2cac1a1197d365e75ecb
+oid sha256:30baefd8be1f50d37d3bc31310171701a7ad8188fdc9a6584558407320f436dc
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d6de7f4a42b33795d5576dbfcbd59d460b4c2829c2dca4e78d73528861030ee0
+oid sha256:591a9c5e45bb1139399d3039b4c05dd34997d5462b11eb1ca1771e8044f4a6a1
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d52a4d67627993645e1cea71ef09276f4b0a2f5bba92959ce9a1a76893653978
+oid sha256:d1351871b6e069a7a4437616d40b913152ecc2cd046e996fcfeb300a3cf60638
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.5751127004623413,
-  "best_model_checkpoint": "./outputs/checkpoint-1300",
-  "epoch": 0.9471766848816029,
+  "best_metric": 1.5539450645446777,
+  "best_model_checkpoint": "./outputs/checkpoint-1400",
+  "epoch": 1.0200364298724955,
   "eval_steps": 100,
-  "global_step": 1300,
+  "global_step": 1400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -189,13 +189,27 @@
       "eval_samples_per_second": 21.638,
       "eval_steps_per_second": 2.707,
       "step": 1300
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.0002,
+      "loss": 1.5538,
+      "step": 1400
+    },
+    {
+      "epoch": 1.02,
+      "eval_loss": 1.5539450645446777,
+      "eval_runtime": 287.8748,
+      "eval_samples_per_second": 21.794,
+      "eval_steps_per_second": 2.727,
+      "step": 1400
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 7.750875343424717e+16,
+  "total_flos": 8.347109656129536e+16,
   "trial_name": null,
   "trial_params": null
 }