Training in progress, step 100

adapter_config.json

@@ -21,19 +21,18 @@
     "ling_emb",
     "pos_enc",
     "attn_norm",
-    "mha",
     "ffn_layer",
     "out_norm",
     "lm_head"
   ],
   "peft_type": "LORA",
-  "r": 64,
+  "r": 32,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "k_proj",
     "q_proj",
-    "v_proj",
-    "k_proj"
+    "v_proj"
   ],
   "task_type": null,
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89676a28495acf20133d07e186a641e2ea25a70f7b809afc29a8c20a0a1f5294
-size 45776132
+oid sha256:a1e397b5d3e0f3cb91c17dabdcc2c8f504242eddcd70e7bf7006f20065ed745e
+size 22170412

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65daa972ffca1926a28354e48760dde9602114de235d9142a48320ec1b4358a7
-size 772305900
+oid sha256:b88839a70db2d293a9c04b775d6c04959c34ba42422bd813e9a0cbb320f3b590
+size 772527092

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae052259db0a8f295fe487cb3f113e7b5fa0b6b1e84a8770a04aa5eb870e5f8c
+oid sha256:5efb7c5c92f60cab9921ebd4e4cf1ba6c0a84597c6f15cfe316d4acc3deeabb9
 size 4920