Training in progress, step 500

config.json

@@ -0,0 +1,63 @@
+{
+  "_name_or_path": "facebook/blenderbot_small-90M",
+  "activation_dropout": 0.0,
+  "activation_function": "gelu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": false,
+  "architectures": [
+    "BlenderbotSmallForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 512,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 2048,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 8,
+  "decoder_start_token_id": 1,
+  "do_blenderbot_90_layernorm": true,
+  "dropout": 0.1,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 2048,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 8,
+  "eos_token_id": 2,
+  "extra_pos_embeddings": 0,
+  "force_bos_token_to_be_generated": false,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_decoder": true,
+  "is_encoder_decoder": false,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "layernorm_variant": "xlm",
+  "length_penalty": 0.65,
+  "max_length": 128,
+  "max_position_embeddings": 512,
+  "min_length": 20,
+  "model_type": "blenderbot-small",
+  "no_repeat_ngram_size": 3,
+  "normalize_before": false,
+  "normalize_embedding": true,
+  "num_beams": 10,
+  "num_hidden_layers": 8,
+  "pad_token_id": 0,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "unk_token_id": 3,
+  "use_cache": true,
+  "vocab_size": 54944
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:044af9fb9bd44ff9c4ff8941a2031b7598cbc7856666528ae10c4118ec3d9450
+size 248131760

runs/Apr26_16-17-42_36e82b9c9f5f/events.out.tfevents.1714148263.36e82b9c9f5f.214.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78c74b3e41087775b9a3e129c745ce7222033508f24bcbf78474239b73d071cf
+size 5659

runs/Apr26_16-19-50_36e82b9c9f5f/events.out.tfevents.1714148391.36e82b9c9f5f.214.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c37f46fef4c1dd5b8b7102105131ac1bfac5a81912e76e5d33d7fa28fc28ff5
+size 5870

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57edf8f17265f31c39592a8cb88e87baf683eb51974bd412f677a118e02b22ad
+size 5048