End of training

README.md

@@ -1,17 +1,19 @@
 ---
 base_model: mistralai/Mistral-7B-Instruct-v0.2
+datasets: sum-pref-overfit
 library_name: transformers
-model_name: prometheus-7b-v1.5-beta-1-0109-sum-pref-overfit-v1-modelcard
+model_name: sum-pref-overfit-v1-modelcard
 tags:
 - generated_from_trainer
+- alignment-handbook
 - trl
 - sft
 licence: license
 ---
 
-# Model Card for prometheus-7b-v1.5-beta-1-0109-sum-pref-overfit-v1-modelcard
+# Model Card for sum-pref-overfit-v1-modelcard
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2).
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the [sum-pref-overfit](https://huggingface.co/datasets/sum-pref-overfit) dataset.
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 1.0,
+    "eval_loss": 0.010753666050732136,
+    "eval_runtime": 1.2774,
+    "eval_samples": 10,
+    "eval_samples_per_second": 0.783,
+    "eval_steps_per_second": 0.783,
     "total_flos": 2.5427771438137344e+16,
     "train_loss": 0.0606694411179249,
     "train_runtime": 598.0457,

config.json

@@ -0,0 +1,28 @@
+{
+  "_attn_implementation_autoset": true,
+  "_name_or_path": "mistralai/Mistral-7B-Instruct-v0.2",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 128,
+  "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": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.3",
+  "use_cache": true,
+  "vocab_size": 32000
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "eval_loss": 0.010753666050732136,
+    "eval_runtime": 1.2774,
+    "eval_samples": 10,
+    "eval_samples_per_second": 0.783,
+    "eval_steps_per_second": 0.783
+}