Upload folder using huggingface_hub

README.md

@@ -206,4 +206,4 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
 [More Information Needed]
 ### Framework versions
 
-- PEFT 0.17.1
+- PEFT 0.18.0

adapter_config.json

@@ -1,9 +1,12 @@
 {
+  "alora_invocation_tokens": null,
   "alpha_pattern": {},
+  "arrow_config": null,
   "auto_mapping": null,
   "base_model_name_or_path": "google/gemma-2-2b",
   "bias": "none",
   "corda_config": null,
+  "ensure_weight_tying": false,
   "eva_config": null,
   "exclude_modules": null,
   "fan_in_fan_out": false,
@@ -20,18 +23,19 @@
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
+  "peft_version": "0.18.0",
   "qalora_group_size": 16,
   "r": 128,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "up_proj",
+    "gate_proj",
     "down_proj",
     "q_proj",
+    "k_proj",
     "o_proj",
-    "v_proj",
-    "gate_proj"
+    "up_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 sha256:942c4e792a20d5d36d62e57ecc20b664777946d0835a9271383afd5e99b85f11
+oid sha256:b06721a9b3d61c6c0c66e2744028ccd466f233ba8b323a55d8f740451ae2c850
 size 664584480

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2373cf17766c2fbe6c76d2c61a20aec8a4ac34fb5d9556819e6fb72699a31531
+oid sha256:2181f8034cbfc2dbfadd605470f68b73ba590be0a8c4032f888499a4f6444e54
 size 1329377575

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:012319d9d7b07efb800bfdc5b30f3b33091204a1f615665fe2368e0bd6978503
+oid sha256:447f6d9c3def923b2023bfae8d2c470e245de58e058e98ae4722cc77fe074f8b
 size 14645

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 100,
-  "best_metric": 0.09553248435258865,
+  "best_metric": 0.09561321139335632,
   "best_model_checkpoint": "/content/models/gemma_qlora_lmh/checkpoint-100",
   "epoch": 1.7008547008547008,
   "eval_steps": 20,
@@ -10,108 +10,108 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "entropy": 2.4642674922943115,
+      "entropy": 2.458172196149826,
       "epoch": 0.3418803418803419,
-      "grad_norm": 6.1703619956970215,
+      "grad_norm": 6.135525226593018,
       "learning_rate": 8.389830508474577e-06,
-      "loss": 0.3828,
-      "mean_token_accuracy": 0.875461021065712,
+      "loss": 0.3827,
+      "mean_token_accuracy": 0.8760345175862312,
       "num_tokens": 113164.0,
       "step": 20
     },
     {
       "epoch": 0.3418803418803419,
-      "eval_entropy": 2.313392945843884,
-      "eval_loss": 0.1408257782459259,
-      "eval_mean_token_accuracy": 0.9526278610922333,
+      "eval_entropy": 2.299590313536489,
+      "eval_loss": 0.14134672284126282,
+      "eval_mean_token_accuracy": 0.9523653464439588,
       "eval_num_tokens": 113164.0,
-      "eval_runtime": 46.6856,
-      "eval_samples_per_second": 39.841,
-      "eval_steps_per_second": 2.506,
+      "eval_runtime": 46.8429,
+      "eval_samples_per_second": 39.707,
+      "eval_steps_per_second": 2.498,
       "step": 20
     },
     {
-      "entropy": 2.3076194286346436,
+      "entropy": 2.298478972911835,
       "epoch": 0.6837606837606838,
-      "grad_norm": 2.0425662994384766,
+      "grad_norm": 2.0751421451568604,
       "learning_rate": 6.694915254237288e-06,
       "loss": 0.1357,
-      "mean_token_accuracy": 0.9569604843854904,
+      "mean_token_accuracy": 0.9575570523738861,
       "num_tokens": 225335.0,
       "step": 40
     },
     {
       "epoch": 0.6837606837606838,
-      "eval_entropy": 2.276767115307669,
-      "eval_loss": 0.1144598051905632,
-      "eval_mean_token_accuracy": 0.9625413275172567,
+      "eval_entropy": 2.2715458065016656,
+      "eval_loss": 0.11509539932012558,
+      "eval_mean_token_accuracy": 0.9629310033260248,
       "eval_num_tokens": 225335.0,
-      "eval_runtime": 45.4774,
-      "eval_samples_per_second": 40.899,
-      "eval_steps_per_second": 2.573,
+      "eval_runtime": 45.696,
+      "eval_samples_per_second": 40.704,
+      "eval_steps_per_second": 2.56,
       "step": 40
     },
     {
-      "entropy": 2.298072344217545,
+      "entropy": 2.295832566725902,
       "epoch": 1.017094017094017,
-      "grad_norm": 2.246678113937378,
+      "grad_norm": 2.188286542892456,
       "learning_rate": 5e-06,
       "loss": 0.113,
-      "mean_token_accuracy": 0.9657873175083063,
+      "mean_token_accuracy": 0.9653458717541817,
       "num_tokens": 330390.0,
       "step": 60
     },
     {
       "epoch": 1.017094017094017,
-      "eval_entropy": 2.2912978331247964,
-      "eval_loss": 0.10871552675962448,
-      "eval_mean_token_accuracy": 0.9649902301975805,
+      "eval_entropy": 2.2908162163873005,
+      "eval_loss": 0.10838180035352707,
+      "eval_mean_token_accuracy": 0.9647057086993487,
       "eval_num_tokens": 330390.0,
-      "eval_runtime": 46.0256,
-      "eval_samples_per_second": 40.412,
-      "eval_steps_per_second": 2.542,
+      "eval_runtime": 46.2535,
+      "eval_samples_per_second": 40.213,
+      "eval_steps_per_second": 2.53,
       "step": 60
     },
     {
-      "entropy": 2.27278618812561,
+      "entropy": 2.271016186475754,
       "epoch": 1.358974358974359,
-      "grad_norm": 2.236058473587036,
+      "grad_norm": 2.2554891109466553,
       "learning_rate": 3.305084745762712e-06,
-      "loss": 0.0845,
-      "mean_token_accuracy": 0.9728620991110801,
+      "loss": 0.0848,
+      "mean_token_accuracy": 0.9718978926539421,
       "num_tokens": 440357.0,
       "step": 80
     },
     {
       "epoch": 1.358974358974359,
-      "eval_entropy": 2.254611888502398,
-      "eval_loss": 0.10490305721759796,
-      "eval_mean_token_accuracy": 0.965580604524694,
+      "eval_entropy": 2.254208923405052,
+      "eval_loss": 0.10406262427568436,
+      "eval_mean_token_accuracy": 0.9654262356269054,
       "eval_num_tokens": 440357.0,
-      "eval_runtime": 46.2372,
-      "eval_samples_per_second": 40.227,
-      "eval_steps_per_second": 2.53,
+      "eval_runtime": 46.3191,
+      "eval_samples_per_second": 40.156,
+      "eval_steps_per_second": 2.526,
       "step": 80
     },
     {
-      "entropy": 2.2653892546892167,
+      "entropy": 2.2658998131752015,
       "epoch": 1.7008547008547008,
-      "grad_norm": 1.7268085479736328,
+      "grad_norm": 1.6946748495101929,
       "learning_rate": 1.6101694915254237e-06,
-      "loss": 0.0715,
-      "mean_token_accuracy": 0.9734208568930626,
+      "loss": 0.0716,
+      "mean_token_accuracy": 0.9734383270144462,
       "num_tokens": 552807.0,
       "step": 100
     },
     {
       "epoch": 1.7008547008547008,
-      "eval_entropy": 2.2389834895093217,
-      "eval_loss": 0.09553248435258865,
-      "eval_mean_token_accuracy": 0.9684329369129279,
+      "eval_entropy": 2.2408512260159874,
+      "eval_loss": 0.09561321139335632,
+      "eval_mean_token_accuracy": 0.9683785734013615,
       "eval_num_tokens": 552807.0,
-      "eval_runtime": 46.1644,
-      "eval_samples_per_second": 40.291,
-      "eval_steps_per_second": 2.534,
+      "eval_runtime": 47.0074,
+      "eval_samples_per_second": 39.568,
+      "eval_steps_per_second": 2.489,
       "step": 100
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8f974810c7f4f0af8e66ac9807b37a99c6690f3fbac636ea7560f6e4b434eb1
+oid sha256:145c7bf7d5850bcddd7a14f18529815a5613136bdd82409c2bf849d5a8d3cdd4
 size 6289