newmindai
/

Llama-3.1-8B-Instruct-w16a8-1node-bs8

Text Generation

mixed-precision

question-answering

distributed-training

text-generation-inference

Model card Files Files and versions

nmmursit commited on 29 days ago

Commit

550c211

·

verified ·

1 Parent(s): 72c11d1

Update training_info.json

Files changed (1) hide show

training_info.json +2 -2

training_info.json CHANGED Viewed

@@ -11,7 +11,7 @@
     "seq_len": 4096,
     "num_samples": null,
     "combine_configs": true,
-    "output_dir": "/gpfs/scratch/ehpc317/esavirdi/startClean/trained_/lama3.1_8b_fp16_fp8",
     "num_epochs": 1,
     "max_steps": 1000,
     "batch_size": 2,
@@ -38,7 +38,7 @@
     "eval_samples": 2000,
     "run_final_eval": true,
     "logging_steps": 1,
-    "log_file": "/gpfs/scratch/ehpc317/esavirdi/startClean/trained_/lama3.1_8b_fp16_fp8/logs/training.log",
     "use_wandb": true,
     "wandb_entity": null,
     "use_tensorboard": false,

     "seq_len": 4096,
     "num_samples": null,
     "combine_configs": true,
+    "output_dir": "./Llama-3.1-8B-Instruct-w16a8-1node",
     "num_epochs": 1,
     "max_steps": 1000,
     "batch_size": 2,
     "eval_samples": 2000,
     "run_final_eval": true,
     "logging_steps": 1,
+    "log_file": "./Llama-3.1-8B-Instruct-w16a8-1node/logs/training.log",
     "use_wandb": true,
     "wandb_entity": null,
     "use_tensorboard": false,