Upload folder using huggingface_hub

pretrained/final/lit_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c9b3dd3d2ba927f4882eff6dffd78d482ccf8342ab9c22aa7641f2a0de220d8
+size 1216478027

pretrained/final/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/final/tokenizer/config.json

@@ -0,0 +1,24 @@
+{
+  "architectures": [
+    "GPTNeoXForCausalLM"
+  ],
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "hidden_act": "gelu",
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neox",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "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
+}

pretrained/final/tokenizer/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|>"
+}

pretrained/latest.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0dc128840c9804e54793cb991e2950b4b12e13f274e8d0b6314a247dba799282
+size 3644179457

pretrained/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00001000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00004000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00005000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00007000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00008000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00010000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00011000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00013000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00014000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00016000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00017000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00019000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00020000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00022000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00023000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00024000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00025000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00026000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00027000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00028000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00029000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00030000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00031000/lit_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4f64ebbc664c532f190e4541cf65147476bd23a94b34a55ef1c754ba69077c3
+size 1216478027

pretrained/step-00031000/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}

pretrained/step-00031052/lit_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c9b3dd3d2ba927f4882eff6dffd78d482ccf8342ab9c22aa7641f2a0de220d8
+size 1216478027

pretrained/step-00031052/model_config.json

@@ -0,0 +1,7 @@
+{
+  "vocab_size": 50304,
+  "seq_len": 1024,
+  "n_layer": 20,
+  "n_embd": 1024,
+  "n_head": 16
+}