Training in progress, step 200

axolotl_config.yaml

@@ -1,5 +1,5 @@
 base_model: Qwen/Qwen2.5-1.5B-Instruct
-batch_size: 32
+batch_size: 40
 bf16: true
 chat_template: tokenizer_default_fallback_alpaca
 datasets:
@@ -15,7 +15,7 @@ datasets:
     system_prompt: ''
 device_map: auto
 eval_sample_packing: false
-eval_steps: 50
+eval_steps: 200
 flash_attention: true
 gradient_checkpointing: true
 group_by_length: true
@@ -24,7 +24,7 @@ hub_strategy: checkpoint
 learning_rate: 0.0002
 logging_steps: 10
 lr_scheduler: cosine
-max_steps: 5000
+max_steps: 10000
 micro_batch_size: 10
 model_type: AutoModelForCausalLM
 num_epochs: 100
@@ -33,11 +33,13 @@ output_dir: /root/.sn56/axolotl/tmp/Qwen2.5-1.5B-Instruct
 pad_to_sequence_len: true
 resize_token_embeddings_to_32x: false
 sample_packing: true
-save_steps: 50
-save_total_limit: 2
+save_steps: 200
+save_total_limit: 1
 sequence_len: 2048
 tokenizer_type: Qwen2TokenizerFast
 torch_dtype: bf16
+training_args_kwargs:
+  hub_private_repo: true
 trust_remote_code: true
 val_set_size: 0.1
 wandb_entity: ''

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57931ccc5ede5e9b408087ee39e0c8acfea478feb0db465454ea28f70f45fce8
+oid sha256:166fb3c7f81e1941b29aa434d5a62e1083fa93f3516c2edd871c14c0ee8ee1e7
 size 3086634632

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:34e809989457dd9bfde4bf2b0182d58f9d3a5925b9457bf37508ace824c39b8b
+oid sha256:6c8cd09cbe4299841bcebbf82901e4817a1a9d47a163f032efe3abc4fb66766c
 size 9144