Training in progress, step 400

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+wandb/run-20240522_054348-vgrzs6jq/run-vgrzs6jq.wandb filter=lfs diff=lfs merge=lfs -text

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:891374b3ee25e1137b95fa1d6d76813663c21e695918d3fcac736bf8ee6bdc5f
+oid sha256:adf2b8d6fdde761c05450f36be5841531cf2f9b29c363ab64000f9d096968d68
 size 1279323952

runs/May22_05-46-00_92dcc4555414/events.out.tfevents.1716356830.92dcc4555414.24.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89bae4605c300c369214b4e1dc6029448770d129b0ef289822025f7ea39bdf88
-size 27897
+oid sha256:98faef80cb9585607e93d29d7a512c5167bb550499e5fa895d9dd7d486f27b88
+size 35517

wandb/run-20240522_054348-vgrzs6jq/files/output.log

@@ -698,3 +698,41 @@ tokenizer config file saved in /kaggle/working/checkpoint-300/tokenizer_config.j
 Special tokens file saved in /kaggle/working/checkpoint-300/special_tokens_map.json
 tokenizer config file saved in /kaggle/working/tokenizer_config.json
 Special tokens file saved in /kaggle/working/special_tokens_map.json
+Deleting older checkpoint [/kaggle/working/checkpoint-200] due to args.save_total_limit
+***** Running Evaluation *****
+  Num examples = 1801
+  Batch size = 1
+Saving model checkpoint to /kaggle/working/checkpoint-400
+loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/config.json
+Model config LlamaConfig {
+  "_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 4096,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.39.3",
+  "use_cache": true,
+  "vocab_size": 32000
+}
+tokenizer config file saved in /kaggle/working/checkpoint-400/tokenizer_config.json
+Special tokens file saved in /kaggle/working/checkpoint-400/special_tokens_map.json
+tokenizer config file saved in /kaggle/working/tokenizer_config.json
+Special tokens file saved in /kaggle/working/special_tokens_map.json

wandb/run-20240522_054348-vgrzs6jq/files/wandb-summary.json

@@ -1 +1 @@
-{"train/loss": 0.4837, "train/grad_norm": 0.0, "train/learning_rate": 1.9760856824352785e-05, "train/rewards/chosen": -20.491928100585938, "train/rewards/rejected": -46.18132019042969, "train/rewards/accuracies": 0.925000011920929, "train/rewards/margins": 25.68939208984375, "train/logps/rejected": -6988.21630859375, "train/logps/chosen": -3619.57568359375, "train/logits/rejected": -3.7832350730895996, "train/logits/chosen": -3.5908398628234863, "train/epoch": 0.17, "train/global_step": 300, "_timestamp": 1716379088.8035662, "_runtime": 22460.168083190918, "_step": 32, "eval/loss": 0.3981175720691681, "eval/runtime": 5068.7089, "eval/samples_per_second": 0.355, "eval/steps_per_second": 0.355, "eval/rewards/chosen": -20.300880432128906, "eval/rewards/rejected": -43.62434768676758, "eval/rewards/accuracies": 0.937812328338623, "eval/rewards/margins": 23.323471069335938, "eval/logps/rejected": -6645.93798828125, "eval/logps/chosen": -3633.760986328125, "eval/logits/rejected": -3.6936769485473633, "eval/logits/chosen": -3.5184967517852783}
+{"train/loss": 0.0197, "train/grad_norm": 0.09612330794334412, "train/learning_rate": 1.9158962821045113e-05, "train/rewards/chosen": -8.07567024230957, "train/rewards/rejected": -30.504592895507812, "train/rewards/accuracies": 1.0, "train/rewards/margins": 22.428918838500977, "train/logps/rejected": -5489.53271484375, "train/logps/chosen": -2377.00732421875, "train/logits/rejected": -3.1316120624542236, "train/logits/chosen": -2.682544231414795, "train/epoch": 0.22, "train/global_step": 400, "_timestamp": 1716386464.2571802, "_runtime": 29835.621697187424, "_step": 43, "eval/loss": 0.17712955176830292, "eval/runtime": 5069.1824, "eval/samples_per_second": 0.355, "eval/steps_per_second": 0.355, "eval/rewards/chosen": -8.881494522094727, "eval/rewards/rejected": -29.159704208374023, "eval/rewards/accuracies": 0.9600222110748291, "eval/rewards/margins": 20.278209686279297, "eval/logps/rejected": -5199.47314453125, "eval/logps/chosen": -2491.822265625, "eval/logits/rejected": -3.1585352420806885, "eval/logits/chosen": -2.733020782470703}