Training in progress, step 6000

adapter_config.json

@@ -1,28 +1,51 @@
 {
-  "alpha_pattern": {},
   "auto_mapping": null,
   "base_model_name_or_path": "gpt2",
-  "bias": "none",
   "fan_in_fan_out": true,
-  "inference_mode": false,
-  "init_lora_weights": true,
-  "layers_pattern": null,
-  "layers_to_transform": null,
-  "loftq_config": {},
-  "lora_alpha": 32,
-  "lora_dropout": 0.01,
-  "megatron_config": null,
-  "megatron_core": "megatron.core",
+  "feedforward_modules": [
+    "transformer.h.0.attn.c_proj",
+    "transformer.h.5.attn.c_proj",
+    "transformer.h.11.attn.c_proj",
+    "transformer.h.9.attn.c_proj",
+    "transformer.h.2.attn.c_proj",
+    "transformer.h.8.attn.c_proj",
+    "transformer.h.10.attn.c_proj",
+    "transformer.h.1.attn.c_proj",
+    "transformer.h.6.attn.c_proj",
+    "transformer.h.3.attn.c_proj",
+    "transformer.h.4.attn.c_proj",
+    "transformer.h.7.attn.c_proj"
+  ],
+  "inference_mode": true,
+  "init_ia3_weights": true,
   "modules_to_save": null,
-  "peft_type": "LORA",
-  "r": 8,
-  "rank_pattern": {},
+  "peft_type": "IA3",
   "revision": null,
   "target_modules": [
-    "c_proj",
-    "c_fc",
-    "c_attn"
+    "transformer.h.3.attn.c_attn",
+    "transformer.h.5.attn.c_proj",
+    "transformer.h.6.attn.c_proj",
+    "transformer.h.4.attn.c_attn",
+    "transformer.h.11.attn.c_proj",
+    "transformer.h.5.attn.c_attn",
+    "transformer.h.8.attn.c_attn",
+    "transformer.h.0.attn.c_attn",
+    "transformer.h.0.attn.c_proj",
+    "transformer.h.6.attn.c_attn",
+    "transformer.h.9.attn.c_proj",
+    "transformer.h.2.attn.c_proj",
+    "transformer.h.1.attn.c_proj",
+    "transformer.h.11.attn.c_attn",
+    "transformer.h.1.attn.c_attn",
+    "transformer.h.7.attn.c_proj",
+    "transformer.h.9.attn.c_attn",
+    "transformer.h.7.attn.c_attn",
+    "transformer.h.8.attn.c_proj",
+    "transformer.h.10.attn.c_proj",
+    "transformer.h.2.attn.c_attn",
+    "transformer.h.3.attn.c_proj",
+    "transformer.h.4.attn.c_proj",
+    "transformer.h.10.attn.c_attn"
   ],
-  "task_type": "CAUSAL_LM",
-  "use_rslora": false
+  "task_type": "CAUSAL_LM"
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e13ef76a1f210827fb843156b1b57d78d50037a4131c82d2a7ad9d458a589960
-size 4730632
+oid sha256:c7a79aafd8a3ded5c8022b7bd591a9b2733484f1509b02860beafed6decdcdff
+size 150264

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a76dc550c0bd6965f6490ca703882eb8539c9f8cddc3d56fb829b9f1e12ef167
-size 4600
+oid sha256:de088ca0b011df1e98d8aba3ad48e1acb1c1af504ac7bc03cf919b6dfb166165
+size 4664