Model save

README.md

@@ -0,0 +1,65 @@
+---
+library_name: peft
+tags:
+- trl
+- sft
+- generated_from_trainer
+base_model: merged_ckpts/skills_red_herring_chat-lora
+model-index:
+- name: skills_red_herring_chat-then-skills_metaphor_chat-lora
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# skills_red_herring_chat-then-skills_metaphor_chat-lora
+
+This model was trained from scratch on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2163
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0002
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 2.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.4383        | 0.96  | 9    | 0.2286          |
+| 0.2047        | 1.92  | 18   | 0.2163          |
+
+
+### Framework versions
+
+- PEFT 0.7.1
+- Transformers 4.37.2
+- Pytorch 2.1.2+cu121
+- Datasets 2.14.6
+- Tokenizers 0.15.1

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 1.92,
+    "eval_loss": 0.21630115807056427,
+    "eval_runtime": 4.1792,
+    "eval_samples": 100,
+    "eval_samples_per_second": 23.928,
+    "eval_steps_per_second": 3.111,
+    "train_loss": 0.2861863639619615,
+    "train_runtime": 207.9549,
+    "train_samples": 300,
+    "train_samples_per_second": 2.885,
+    "train_steps_per_second": 0.087
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.92,
+    "eval_loss": 0.21630115807056427,
+    "eval_runtime": 4.1792,
+    "eval_samples": 100,
+    "eval_samples_per_second": 23.928,
+    "eval_steps_per_second": 3.111
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.92,
+    "train_loss": 0.2861863639619615,
+    "train_runtime": 207.9549,
+    "train_samples": 300,
+    "train_samples_per_second": 2.885,
+    "train_steps_per_second": 0.087
+}

trainer_state.json

@@ -0,0 +1,70 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.92,
+  "eval_steps": 500,
+  "global_step": 18,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0001,
+      "loss": 0.6469,
+      "step": 1
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00018314696123025454,
+      "loss": 0.4383,
+      "step": 5
+    },
+    {
+      "epoch": 0.96,
+      "eval_loss": 0.2286338359117508,
+      "eval_runtime": 5.2482,
+      "eval_samples_per_second": 19.054,
+      "eval_steps_per_second": 2.477,
+      "step": 9
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0001,
+      "loss": 0.23,
+      "step": 10
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.6853038769745467e-05,
+      "loss": 0.2047,
+      "step": 15
+    },
+    {
+      "epoch": 1.92,
+      "eval_loss": 0.21630115807056427,
+      "eval_runtime": 4.1963,
+      "eval_samples_per_second": 23.83,
+      "eval_steps_per_second": 3.098,
+      "step": 18
+    },
+    {
+      "epoch": 1.92,
+      "step": 18,
+      "total_flos": 19174999326720.0,
+      "train_loss": 0.2861863639619615,
+      "train_runtime": 207.9549,
+      "train_samples_per_second": 2.885,
+      "train_steps_per_second": 0.087
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 18,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "total_flos": 19174999326720.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}