End of training

README.md

@@ -1,8 +1,11 @@
 ---
 base_model: alignment-handbook/zephyr-7b-sft-full
+datasets:
+- SeniorKabanocci/oasst1-binarized
 library_name: peft
 license: apache-2.0
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
@@ -17,7 +20,17 @@ should probably proofread and complete it, then remove this comment. -->
 [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/jan-majkutewicz/huggingface/runs/uistaj1t)
 # zephyr-7b-oasst1
 
-This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on an unknown dataset.
+This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the SeniorKabanocci/oasst1-binarized dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.6059
+- Rewards/chosen: -0.9430
+- Rewards/rejected: -1.3459
+- Rewards/accuracies: 0.6348
+- Rewards/margins: 0.4030
+- Logps/rejected: -239.4801
+- Logps/chosen: -256.8972
+- Logits/rejected: -2.8973
+- Logits/chosen: -2.9273
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9994353472614342,
+    "eval_logits/chosen": -2.9273033142089844,
+    "eval_logits/rejected": -2.8973140716552734,
+    "eval_logps/chosen": -256.8972473144531,
+    "eval_logps/rejected": -239.4800567626953,
+    "eval_loss": 0.6058722734451294,
+    "eval_rewards/accuracies": 0.6348314881324768,
+    "eval_rewards/chosen": -0.9429696798324585,
+    "eval_rewards/margins": 0.4029598534107208,
+    "eval_rewards/rejected": -1.3459293842315674,
+    "eval_runtime": 230.8014,
+    "eval_samples": 712,
+    "eval_samples_per_second": 3.085,
+    "eval_steps_per_second": 0.386,
     "total_flos": 0.0,
     "train_loss": 0.6616038349388683,
     "train_runtime": 9964.5835,

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "alignment-handbook/zephyr-7b-sft-full",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.42.3",
+  "use_cache": true,
+  "vocab_size": 32000
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 0.9994353472614342,
+    "eval_logits/chosen": -2.9273033142089844,
+    "eval_logits/rejected": -2.8973140716552734,
+    "eval_logps/chosen": -256.8972473144531,
+    "eval_logps/rejected": -239.4800567626953,
+    "eval_loss": 0.6058722734451294,
+    "eval_rewards/accuracies": 0.6348314881324768,
+    "eval_rewards/chosen": -0.9429696798324585,
+    "eval_rewards/margins": 0.4029598534107208,
+    "eval_rewards/rejected": -1.3459293842315674,
+    "eval_runtime": 230.8014,
+    "eval_samples": 712,
+    "eval_samples_per_second": 3.085,
+    "eval_steps_per_second": 0.386
+}