update eval yaml

scripts/eval/yamls/long_context_eval_8k.yaml

@@ -2,28 +2,32 @@ max_seq_len: 8192
 seed: 1
 precision: amp_bf16
 
+
+# If you are using one model, put it here:
+model_name_or_path: EleutherAI/gpt-neo-125m
+# model_name_or_path: mosaicml/mpt-7b-chat-8k
+# otherwise, write a block for each model you want to test in the `models` section
+
 models:
 -
-  model_name: mosaicml/mpt-7b-chat-8k
+  model_name: ${model_name_or_path}
   model:
     name: hf_causal_lm
-    pretrained_model_name_or_path: mosaicml/mpt-7b-chat-8k
+    pretrained_model_name_or_path: ${model_name_or_path}
+    init_device: cpu
     pretrained: true
-    attn_config:
-      attn_impl: triton
   tokenizer:
-    name: mosaicml/mpt-7b-chat-8k
+    name: ${model_name_or_path}
     kwargs:
       model_max_length: ${max_seq_len}
 
 device_eval_batch_size: 1 
 
-# FSDP config for model sharding
 fsdp_config:
   sharding_strategy: FULL_SHARD
   mixed_precision: FULL
 
-icl_tasks: 'eval/yamls/leval.yaml'
-eval_gauntlet: 'eval/yamls/leval_test.yaml'
-# icl_tasks: 'eval/yamls/long_context_tasks.yaml'
-# eval_gauntlet: 'eval/yamls/eval_gauntlet_8k_section.yaml'
+# icl_tasks: 'eval/yamls/leval.yaml'
+# eval_gauntlet: 'eval/yamls/leval_test.yaml'
+icl_tasks: 'eval/yamls/long_context_tasks.yaml'
+eval_gauntlet: 'eval/yamls/eval_gauntlet_8k_section.yaml'