Upload folder using huggingface_hub

config.json

@@ -0,0 +1,24 @@
+{
+  "architectures": [
+    "GPTNeoXForCausalLM"
+  ],
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "hidden_act": "gelu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neox",
+  "num_attention_heads": 8,
+  "num_hidden_layers": 16,
+  "rotary_emb_base": 10000,
+  "rotary_pct": 0.25,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.24.0",
+  "use_cache": true,
+  "use_parallel_residual": true,
+  "vocab_size": 50304
+}

hyperparameters.yaml

@@ -0,0 +1,37 @@
+model_name: pythia-1b
+out_dir: /data/users/zichunyu/out/pythia-1b/fineweb/sample-100BT
+resume: false
+data:
+  class_path: litgpt.data.FineWeb
+  init_args:
+    data_path: /data/users/zichunyu/data/fineweb/sample-100BT
+    val_split_fraction: 0.0005
+    seed: 42
+    num_workers: 8
+train:
+  save_interval: 5000
+  log_interval: 50
+  global_batch_size: 512
+  micro_batch_size: 16
+  lr_warmup_steps: 2000
+  max_tokens: 50000000000
+  tie_embeddings: false
+  max_norm: 1.0
+  min_lr: 4.0e-05
+eval:
+  interval: 200000
+  max_iters: 100
+  initial_validation: false
+optimizer:
+  class_path: torch.optim.AdamW
+  init_args:
+    lr: 0.0004
+    weight_decay: 0.1
+    betas:
+    - 0.9
+    - 0.95
+devices: auto
+tokenizer_dir: checkpoints/EleutherAI/pythia-1b
+logger_name: wandb
+exp_name: pythia-1b_fineweb_sample-100BT
+seed: 42

lit_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66aee03d68a422cf10f58a5d25852defd931eb96838e75b702003de8b2b45945
+size 12141642522

model_config.yaml

@@ -0,0 +1,28 @@
+bias: true
+block_size: 2048
+gelu_approximate: none
+head_size: 256
+hf_config:
+  name: pythia-1b
+  org: EleutherAI
+intermediate_size: 8192
+lm_head_bias: false
+mlp_class_name: GptNeoxMLP
+n_embd: 2048
+n_expert: 0
+n_expert_per_token: 0
+n_head: 8
+n_layer: 16
+n_query_groups: 8
+name: pythia-1b
+norm_class_name: LayerNorm
+norm_eps: 1.0e-05
+padded_vocab_size: 50304
+padding_multiple: 128
+parallel_residual: true
+rope_base: 10000
+rope_condense_ratio: 1
+rotary_percentage: 0.25
+scale_embeddings: false
+shared_attention_norm: false
+vocab_size: 50254

tokenizer_config.json

@@ -0,0 +1,9 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "name_or_path": "EleutherAI/gpt-neox-20b",
+  "special_tokens_map_file": "/admin/home-hailey/.cache/huggingface/hub/models--EleutherAI--gpt-neox-20b/snapshots/4e49eadb5d14bd22f314ec3f45b69a87b88c7691/special_tokens_map.json",
+  "tokenizer_class": "GPTNeoXTokenizer",
+  "unk_token": "<|endoftext|>"
+}