Training in progress, step 20

adapter_config.json

@@ -21,9 +21,9 @@
   "revision": null,
   "target_modules": [
     "query_key_value",
-    "dense_h_to_4h",
+    "dense",
     "dense_4h_to_h",
-    "dense"
+    "dense_h_to_4h"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af4430b06ece65f9158c8afbfaf651268fdeff15a8b2dd4f1f9ef165d5657c80
+oid sha256:dc24c98fdb5b0f5b2c2d099df00de37c74c630cdb923aa00e4188749ddb432a2
 size 1579384

config.json

@@ -20,21 +20,6 @@
   "num_attention_heads": 4,
   "num_hidden_layers": 6,
   "partial_rotary_factor": 0.25,
-  "quantization_config": {
-    "_load_in_4bit": true,
-    "_load_in_8bit": false,
-    "bnb_4bit_compute_dtype": "bfloat16",
-    "bnb_4bit_quant_storage": "bfloat16",
-    "bnb_4bit_quant_type": "nf4",
-    "bnb_4bit_use_double_quant": true,
-    "llm_int8_enable_fp32_cpu_offload": false,
-    "llm_int8_has_fp16_weight": false,
-    "llm_int8_skip_modules": null,
-    "llm_int8_threshold": 6.0,
-    "load_in_4bit": true,
-    "load_in_8bit": false,
-    "quant_method": "bitsandbytes"
-  },
   "rope_scaling": null,
   "rope_theta": 10000,
   "rotary_emb_base": 10000,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3a10b23459776d70a64cd0caca7249c57668b0e560e1f8f4ee49ded5105bf99
+oid sha256:3c59b827379aab605d7eebc5971fb311c91841c53277e301626e1a492c348a7a
 size 6776