End of training

README.md

@@ -3,11 +3,13 @@ library_name: peft
 license: apache-2.0
 base_model: HuggingFaceTB/SmolLM2-1.7B-Instruct
 tags:
+- alignment-handbook
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- HuggingFaceH4/cai-conversation-harmless
+- HuggingFaceH4/ultrachat_200k
 model-index:
 - name: smollm2-1.7b-instr-sft-cai
   results: []
@@ -18,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # smollm2-1.7b-instr-sft-cai
 
-This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-1.7B-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B-Instruct) on the generator dataset.
+This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-1.7B-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B-Instruct) on the HuggingFaceH4/cai-conversation-harmless and the HuggingFaceH4/ultrachat_200k datasets.
 It achieves the following results on the evaluation set:
 - Loss: 1.2565
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 0.9913258983890955,
+    "eval_loss": 1.2565181255340576,
+    "eval_runtime": 3317.262,
+    "eval_samples": 24265,
+    "eval_samples_per_second": 4.291,
+    "eval_steps_per_second": 0.537,
     "total_flos": 1.280941246906368e+17,
     "train_loss": 1.281534285545349,
     "train_runtime": 8293.439,

config.json

@@ -0,0 +1,53 @@
+{
+  "_attn_implementation_autoset": true,
+  "_name_or_path": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 8192,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 32,
+  "pad_token_id": 2,
+  "pretraining_tp": 1,
+  "quantization_config": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
+    "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_quant_storage": "uint8",
+    "bnb_4bit_quant_type": "nf4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 130000,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers.js_config": {
+    "kv_cache_dtype": {
+      "fp16": "float16",
+      "q4f16": "float16"
+    }
+  },
+  "transformers_version": "4.47.1",
+  "use_cache": true,
+  "vocab_size": 49152
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.9913258983890955,
+    "eval_loss": 1.2565181255340576,
+    "eval_runtime": 3317.262,
+    "eval_samples": 24265,
+    "eval_samples_per_second": 4.291,
+    "eval_steps_per_second": 0.537
+}

runs/Jan03_10-24-26_pkaiju/events.out.tfevents.1735911492.pkaiju.20743.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7dc1421756ff08de2c515dc9c1f535a3f7e7494793646e393d8b40a2fb8ac9c6
+size 354