Training in progress, step 100

axolotl_config.yaml

@@ -1,5 +1,5 @@
 base_model: trl-internal-testing/tiny-random-LlamaForCausalLM
-batch_size: 256
+batch_size: 128
 bf16: true
 chat_template: tokenizer_default_fallback_alpaca
 datasets:
@@ -24,7 +24,7 @@ hub_strategy: checkpoint
 learning_rate: 0.0002
 logging_steps: 10
 lr_scheduler: cosine
-max_steps: 20000
+max_steps: 10000
 micro_batch_size: 32
 model_type: AutoModelForCausalLM
 num_epochs: 100
@@ -38,6 +38,8 @@ save_total_limit: 1
 sequence_len: 2048
 tokenizer_type: LlamaTokenizerFast
 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:ae2287b45658d1cae69b4bfe25e777f022036e8d980148fa7635d454487588a9
+oid sha256:25a257c5b2e34eda80e3d84f1a8cc4247ba163d63b057609915a20d5c03487fe
 size 2066752

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9629b97ec393ef093e4840cb93f3f7c6eafc6cfa1b3cb229133bf54718ca98cb
+oid sha256:11cf48786efac37806223cc8882d4253d84ff8c3599dc92c9fda0e12bc8a651f
 size 6840