End of training

README.md

@@ -2,9 +2,15 @@
 license: apache-2.0
 base_model: martimfasantos/tinyllama-1.1b-sum-sft-full_old
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- openai/summarize_from_feedback
 model-index:
 - name: tinyllama-1.1b-sum-dpo-full_LR2e-7_3epochs_old
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tinyllama-1.1b-sum-dpo-full_LR2e-7_3epochs_old
 
-This model is a fine-tuned version of [martimfasantos/tinyllama-1.1b-sum-sft-full_old](https://huggingface.co/martimfasantos/tinyllama-1.1b-sum-sft-full_old) on an unknown dataset.
+This model is a fine-tuned version of [martimfasantos/tinyllama-1.1b-sum-sft-full_old](https://huggingface.co/martimfasantos/tinyllama-1.1b-sum-sft-full_old) on the openai/summarize_from_feedback dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6303
-- Rewards/chosen: -1.4484
-- Rewards/rejected: -1.8080
-- Rewards/accuracies: 0.6436
-- Rewards/margins: 0.3596
-- Logps/rejected: -243.9776
-- Logps/chosen: -203.5508
-- Logits/rejected: -1.7024
-- Logits/chosen: -1.7262
+- Loss: 0.6307
+- Rewards/chosen: -1.4504
+- Rewards/rejected: -1.8097
+- Rewards/accuracies: 0.6434
+- Rewards/margins: 0.3593
+- Logps/rejected: -244.1550
+- Logps/chosen: -203.7530
+- Logits/rejected: -1.7026
+- Logits/chosen: -1.7263
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 3.0,
+    "eval_logits/chosen": -1.726296067237854,
+    "eval_logits/rejected": -1.7025630474090576,
+    "eval_logps/chosen": -203.75299072265625,
+    "eval_logps/rejected": -244.15496826171875,
+    "eval_loss": 0.6307257413864136,
+    "eval_rewards/accuracies": 0.6433550119400024,
+    "eval_rewards/chosen": -1.4504109621047974,
+    "eval_rewards/margins": 0.3593376576900482,
+    "eval_rewards/rejected": -1.809748649597168,
+    "eval_runtime": 384.3784,
+    "eval_samples": 4304,
+    "eval_samples_per_second": 11.197,
+    "eval_steps_per_second": 1.4,
     "total_flos": 0.0,
     "train_loss": 0.5724969553024556,
     "train_runtime": 86264.9537,

config.json

@@ -24,6 +24,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "float32",
   "transformers_version": "4.41.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 3.0,
+    "eval_logits/chosen": -1.726296067237854,
+    "eval_logits/rejected": -1.7025630474090576,
+    "eval_logps/chosen": -203.75299072265625,
+    "eval_logps/rejected": -244.15496826171875,
+    "eval_loss": 0.6307257413864136,
+    "eval_rewards/accuracies": 0.6433550119400024,
+    "eval_rewards/chosen": -1.4504109621047974,
+    "eval_rewards/margins": 0.3593376576900482,
+    "eval_rewards/rejected": -1.809748649597168,
+    "eval_runtime": 384.3784,
+    "eval_samples": 4304,
+    "eval_samples_per_second": 11.197,
+    "eval_steps_per_second": 1.4
+}

runs/Jun11_00-56-54_poseidon/events.out.tfevents.1718154602.poseidon.4172683.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a30564956da39e11c00772aff16ec8db961ebd9ddc3bf476b81006281999dc1
+size 841