Training in progress, step 1000

adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": null,
+  "inference_mode": false,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "k_proj",
+    "o_proj",
+    "down_proj",
+    "gate_proj",
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

config.json

@@ -20,7 +20,7 @@
   "sliding_window": 32768,
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.51.0",
+  "transformers_version": "4.51.3",
   "use_cache": false,
   "use_sliding_window": false,
   "vocab_size": 151936

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:88ffaf399c01d63fc9f1335a19c1d4d47f650c42e7c59108a136b71ae481194a
+oid sha256:77484cafe0ac030d4a24440e079daed0a0f20943974fb667686f912f8948f0e5
 size 3087467144

runs/Apr17_15-18-53_dws-13.informatik.uni-mannheim.de/events.out.tfevents.1744895946.dws-13.informatik.uni-mannheim.de.346181.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5dca305dd35e6ad8052f1ca286aa7fc351bc2c91059df391a82a74b0ea90e939
+size 7279

runs/Apr17_15-21-20_dws-13.informatik.uni-mannheim.de/events.out.tfevents.1744896097.dws-13.informatik.uni-mannheim.de.347590.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46e60667aa09af54b351db133bc37245f95900c3cb60cf948f38a7c3924fa6fb
+size 16799

runs/Apr17_15-30-37_dws-13.informatik.uni-mannheim.de/events.out.tfevents.1744896656.dws-13.informatik.uni-mannheim.de.352082.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05c2258c915174c2689550bfad8f1bde5cb07395e14ec388fbe78d9f96afbe7b
+size 14080

runs/Apr17_15-34-47_dws-13.informatik.uni-mannheim.de/events.out.tfevents.1744896907.dws-13.informatik.uni-mannheim.de.354277.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3027cb401193747296d7cecab4181c7ff1b8f4cacc384466da95dbafc6237f88
+size 10000

runs/Apr17_15-38-37_dws-13.informatik.uni-mannheim.de/events.out.tfevents.1744897136.dws-13.informatik.uni-mannheim.de.358574.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3b1a32ef20678274e49c312b06c745d9f88db7ad8df82b237dd4e6b404656e1
+size 15440

runs/Apr17_15-44-21_dws-13.informatik.uni-mannheim.de/events.out.tfevents.1744897478.dws-13.informatik.uni-mannheim.de.361911.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01457b5632ba720ae0fbd37c68146b6df655770b9375811ca122e7247b375e08
+size 10680

runs/Apr17_15-47-02_dws-13.informatik.uni-mannheim.de/events.out.tfevents.1744897641.dws-13.informatik.uni-mannheim.de.367715.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16d7eba0e2149a84b1d8c4354de5c722f374f8261f90dea6a210f67c99553175
+size 73239

runs/Apr17_17-06-05_dws-13.informatik.uni-mannheim.de/events.out.tfevents.1744902382.dws-13.informatik.uni-mannheim.de.397931.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5fb65086de7896a9568fa730ba3a18b970385091a6e1f2d29e4936cdef3d9ae
+size 697781

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:222bc846b2070abf618f8e8b27e62457d02026c1547537c819cda73895175bc8
-size 7608
+oid sha256:f54aab527e52224d27152bc9919e57e2e695d6220dc31b2172ed3380d9ca2eca
+size 7672