Trained for 5 epochs

config.json

@@ -0,0 +1,42 @@
+{
+  "_name_or_path": "Salesforce/codegen-350M-mono",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "CodeGenForCausalLM"
+  ],
+  "attn_pdrop": 0.0,
+  "bos_token_id": 1,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 50256,
+  "gradient_checkpointing": false,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "codegen",
+  "n_ctx": 2048,
+  "n_embd": 1024,
+  "n_head": 16,
+  "n_inner": null,
+  "n_layer": 20,
+  "n_positions": 2048,
+  "resid_pdrop": 0.0,
+  "rotary_dim": 32,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50,
+      "temperature": 1.0
+    }
+  },
+  "tie_word_embeddings": false,
+  "tokenizer_class": "GPT2Tokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.31.0",
+  "use_cache": true,
+  "vocab_size": 51200
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 50256,
+  "transformers_version": "4.31.0"
+}

pytorch_model-00001-of-00009.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:491354708b3a4a475588350611d6baf4b087d8d978c7d36e128095dcc0c8da58
+size 209716138

pytorch_model-00002-of-00009.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81d9b906272fb4110704b5742bfb567571b9c6041b9951769f4c8458bc059e7e
+size 184669675

pytorch_model-00003-of-00009.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e821ce2de1a5a6a4ad68b236f36b8d4ff052cd5024cef722c013c8f5f9854b47
+size 184657451

pytorch_model-00004-of-00009.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4bf2d192b5b3c549057a96ad5036d3b9b46f1ff665681b8ee197314a88f2cad
+size 197257021

pytorch_model-00005-of-00009.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f8fd7738191350d6655a039ef9276d6a00ce9032998026d566ade2980ff2de1
+size 188868413

pytorch_model-00006-of-00009.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11cf7d1c5830af666bf15f1fa639fe9e551cd44f7cd94e7f21ea1409fd3da7be
+size 184660935

pytorch_model-00007-of-00009.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e3c43019eb487e36cba72c5c8b3e2e7bbaaaa126397aa2f785e65094af7ee15
+size 67154087

pytorch_model-00008-of-00009.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19c27e2f8ad8724831e7e9b2f77c448a83a247d6835c6cd3a7304037daf6e51c
+size 209716138

pytorch_model-00009-of-00009.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efcdabaa0e93a3794744118162be068235917eec96747a399cd995e1f527d63b
+size 205738

pytorch_model.bin.index.json

@@ -0,0 +1,172 @@
+{
+  "metadata": {
+    "total_size": 1426849792
+  },
+  "weight_map": {
+    "lm_head.bias": "pytorch_model-00009-of-00009.bin",
+    "lm_head.weight": "pytorch_model-00008-of-00009.bin",
+    "transformer.h.0.attn.out_proj.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.0.attn.qkv_proj.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.0.ln_1.bias": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.0.ln_1.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.0.mlp.fc_in.bias": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.0.mlp.fc_in.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.0.mlp.fc_out.bias": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.0.mlp.fc_out.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.1.attn.out_proj.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.1.attn.qkv_proj.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.1.ln_1.bias": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.1.ln_1.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.1.mlp.fc_in.bias": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.1.mlp.fc_in.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.1.mlp.fc_out.bias": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.1.mlp.fc_out.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.10.attn.out_proj.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.10.attn.qkv_proj.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.10.ln_1.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.10.ln_1.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.10.mlp.fc_in.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.10.mlp.fc_in.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.10.mlp.fc_out.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.10.mlp.fc_out.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.11.attn.out_proj.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.11.attn.qkv_proj.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.11.ln_1.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.11.ln_1.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.11.mlp.fc_in.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.11.mlp.fc_in.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.11.mlp.fc_out.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.11.mlp.fc_out.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.12.attn.out_proj.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.12.attn.qkv_proj.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.12.ln_1.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.12.ln_1.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.12.mlp.fc_in.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.12.mlp.fc_in.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.12.mlp.fc_out.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.12.mlp.fc_out.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.13.attn.out_proj.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.13.attn.qkv_proj.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.13.ln_1.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.13.ln_1.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.13.mlp.fc_in.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.13.mlp.fc_in.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.13.mlp.fc_out.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.13.mlp.fc_out.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.14.attn.out_proj.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.14.attn.qkv_proj.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.14.ln_1.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.14.ln_1.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.14.mlp.fc_in.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.14.mlp.fc_in.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.14.mlp.fc_out.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.14.mlp.fc_out.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.15.attn.out_proj.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.15.attn.qkv_proj.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.15.ln_1.bias": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.15.ln_1.weight": "pytorch_model-00005-of-00009.bin",
+    "transformer.h.15.mlp.fc_in.bias": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.15.mlp.fc_in.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.15.mlp.fc_out.bias": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.15.mlp.fc_out.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.16.attn.out_proj.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.16.attn.qkv_proj.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.16.ln_1.bias": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.16.ln_1.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.16.mlp.fc_in.bias": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.16.mlp.fc_in.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.16.mlp.fc_out.bias": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.16.mlp.fc_out.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.17.attn.out_proj.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.17.attn.qkv_proj.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.17.ln_1.bias": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.17.ln_1.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.17.mlp.fc_in.bias": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.17.mlp.fc_in.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.17.mlp.fc_out.bias": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.17.mlp.fc_out.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.18.attn.out_proj.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.18.attn.qkv_proj.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.18.ln_1.bias": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.18.ln_1.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.18.mlp.fc_in.bias": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.18.mlp.fc_in.weight": "pytorch_model-00006-of-00009.bin",
+    "transformer.h.18.mlp.fc_out.bias": "pytorch_model-00007-of-00009.bin",
+    "transformer.h.18.mlp.fc_out.weight": "pytorch_model-00007-of-00009.bin",
+    "transformer.h.19.attn.out_proj.weight": "pytorch_model-00007-of-00009.bin",
+    "transformer.h.19.attn.qkv_proj.weight": "pytorch_model-00007-of-00009.bin",
+    "transformer.h.19.ln_1.bias": "pytorch_model-00007-of-00009.bin",
+    "transformer.h.19.ln_1.weight": "pytorch_model-00007-of-00009.bin",
+    "transformer.h.19.mlp.fc_in.bias": "pytorch_model-00007-of-00009.bin",
+    "transformer.h.19.mlp.fc_in.weight": "pytorch_model-00007-of-00009.bin",
+    "transformer.h.19.mlp.fc_out.bias": "pytorch_model-00007-of-00009.bin",
+    "transformer.h.19.mlp.fc_out.weight": "pytorch_model-00007-of-00009.bin",
+    "transformer.h.2.attn.out_proj.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.2.attn.qkv_proj.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.2.ln_1.bias": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.2.ln_1.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.2.mlp.fc_in.bias": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.2.mlp.fc_in.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.2.mlp.fc_out.bias": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.2.mlp.fc_out.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.3.attn.out_proj.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.3.attn.qkv_proj.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.3.ln_1.bias": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.3.ln_1.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.3.mlp.fc_in.bias": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.3.mlp.fc_in.weight": "pytorch_model-00002-of-00009.bin",
+    "transformer.h.3.mlp.fc_out.bias": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.3.mlp.fc_out.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.4.attn.out_proj.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.4.attn.qkv_proj.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.4.ln_1.bias": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.4.ln_1.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.4.mlp.fc_in.bias": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.4.mlp.fc_in.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.4.mlp.fc_out.bias": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.4.mlp.fc_out.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.5.attn.out_proj.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.5.attn.qkv_proj.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.5.ln_1.bias": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.5.ln_1.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.5.mlp.fc_in.bias": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.5.mlp.fc_in.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.5.mlp.fc_out.bias": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.5.mlp.fc_out.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.6.attn.out_proj.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.6.attn.qkv_proj.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.6.ln_1.bias": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.6.ln_1.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.6.mlp.fc_in.bias": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.6.mlp.fc_in.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.6.mlp.fc_out.bias": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.6.mlp.fc_out.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.7.attn.out_proj.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.7.attn.qkv_proj.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.7.ln_1.bias": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.7.ln_1.weight": "pytorch_model-00003-of-00009.bin",
+    "transformer.h.7.mlp.fc_in.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.7.mlp.fc_in.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.7.mlp.fc_out.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.7.mlp.fc_out.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.8.attn.out_proj.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.8.attn.qkv_proj.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.8.ln_1.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.8.ln_1.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.8.mlp.fc_in.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.8.mlp.fc_in.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.8.mlp.fc_out.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.8.mlp.fc_out.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.9.attn.out_proj.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.9.attn.qkv_proj.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.9.ln_1.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.9.ln_1.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.9.mlp.fc_in.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.9.mlp.fc_in.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.9.mlp.fc_out.bias": "pytorch_model-00004-of-00009.bin",
+    "transformer.h.9.mlp.fc_out.weight": "pytorch_model-00004-of-00009.bin",
+    "transformer.ln_f.bias": "pytorch_model-00007-of-00009.bin",
+    "transformer.ln_f.weight": "pytorch_model-00007-of-00009.bin",
+    "transformer.wte.weight": "pytorch_model-00001-of-00009.bin"
+  }
+}