gpt2-1b-miner-3 / config.json
kmfoda's picture
Run 5. Outer Step 59. Inner Step 65.
521f73f verified
{
"_name_or_path": "distributed/llama-1b",
"all_reduce_scores": {
"0": "NON_PARTICIPATING",
"1": "SUCCESS",
"10": "SUCCESS",
"100": "NON_PARTICIPATING",
"101": "SUCCESS",
"102": "NON_PARTICIPATING",
"103": "SUCCESS",
"104": "SUCCESS",
"105": "SUCCESS",
"106": "NON_PARTICIPATING",
"107": "SUCCESS",
"108": "NON_PARTICIPATING",
"109": "SUCCESS",
"11": "SUCCESS",
"110": "NON_PARTICIPATING",
"111": "SUCCESS",
"112": "SUCCESS",
"113": "NON_PARTICIPATING",
"114": "NON_PARTICIPATING",
"115": "SUCCESS",
"116": "NON_PARTICIPATING",
"117": "SUCCESS",
"118": "SUCCESS",
"119": "NON_PARTICIPATING",
"12": "SUCCESS",
"120": "SUCCESS",
"121": "NON_PARTICIPATING",
"122": "NON_PARTICIPATING",
"123": "SUCCESS",
"124": "SUCCESS",
"125": "SUCCESS",
"126": "SUCCESS",
"127": "NON_PARTICIPATING",
"128": "SUCCESS",
"129": "NON_PARTICIPATING",
"13": "NON_PARTICIPATING",
"130": "NON_PARTICIPATING",
"131": "NON_PARTICIPATING",
"132": "NON_PARTICIPATING",
"133": "NON_PARTICIPATING",
"134": "SUCCESS",
"135": "SUCCESS",
"136": "NON_PARTICIPATING",
"137": "SUCCESS",
"138": "SUCCESS",
"139": "SUCCESS",
"14": "SUCCESS",
"140": "SUCCESS",
"141": "NON_PARTICIPATING",
"142": "NON_PARTICIPATING",
"143": "SUCCESS",
"144": "NON_PARTICIPATING",
"145": "NON_PARTICIPATING",
"146": "SUCCESS",
"147": "SUCCESS",
"148": "SUCCESS",
"149": "SUCCESS",
"15": "SUCCESS",
"150": "NON_PARTICIPATING",
"151": "NON_PARTICIPATING",
"152": "NON_PARTICIPATING",
"153": "NON_PARTICIPATING",
"154": "NON_PARTICIPATING",
"155": "SUCCESS",
"156": "NON_PARTICIPATING",
"157": "NON_PARTICIPATING",
"158": "NON_PARTICIPATING",
"159": "NON_PARTICIPATING",
"16": "SUCCESS",
"160": "SUCCESS",
"161": "NON_PARTICIPATING",
"162": "NON_PARTICIPATING",
"163": "NON_PARTICIPATING",
"164": "SUCCESS",
"165": "NON_PARTICIPATING",
"166": "SUCCESS",
"167": "NON_PARTICIPATING",
"168": "NON_PARTICIPATING",
"169": "NON_PARTICIPATING",
"17": "SUCCESS",
"170": "SUCCESS",
"171": "SUCCESS",
"172": "NON_PARTICIPATING",
"173": "SUCCESS",
"174": "SUCCESS",
"175": "NON_PARTICIPATING",
"176": "NON_PARTICIPATING",
"177": "SUCCESS",
"178": "SUCCESS",
"179": "NON_PARTICIPATING",
"18": "NON_PARTICIPATING",
"180": "SUCCESS",
"181": "SUCCESS",
"182": "SUCCESS",
"183": "SUCCESS",
"184": "SUCCESS",
"185": "SUCCESS",
"186": "NON_PARTICIPATING",
"187": "SUCCESS",
"188": "SUCCESS",
"189": "SUCCESS",
"19": "SUCCESS",
"190": "NON_PARTICIPATING",
"191": "NON_PARTICIPATING",
"192": "SUCCESS",
"193": "SUCCESS",
"194": "SUCCESS",
"195": "NON_PARTICIPATING",
"196": "NON_PARTICIPATING",
"197": "NON_PARTICIPATING",
"198": "SUCCESS",
"199": "NON_PARTICIPATING",
"2": "SUCCESS",
"20": "SUCCESS",
"200": "SUCCESS",
"201": "NON_PARTICIPATING",
"202": "SUCCESS",
"203": "SUCCESS",
"204": "SUCCESS",
"205": "NON_PARTICIPATING",
"206": "NON_PARTICIPATING",
"207": "NON_PARTICIPATING",
"208": "NON_PARTICIPATING",
"209": "NON_PARTICIPATING",
"21": "SUCCESS",
"210": "SUCCESS",
"211": "SUCCESS",
"212": "NON_PARTICIPATING",
"213": "NON_PARTICIPATING",
"214": "NON_PARTICIPATING",
"215": "NON_PARTICIPATING",
"216": "SUCCESS",
"217": "SUCCESS",
"218": "SUCCESS",
"219": "SUCCESS",
"22": "SUCCESS",
"220": "SUCCESS",
"221": "NON_PARTICIPATING",
"222": "NON_PARTICIPATING",
"223": "SUCCESS",
"224": "NON_PARTICIPATING",
"225": "NON_PARTICIPATING",
"226": "NON_PARTICIPATING",
"227": "SUCCESS",
"228": "NON_PARTICIPATING",
"229": "NON_PARTICIPATING",
"23": "SUCCESS",
"230": "FAIL",
"231": "NON_PARTICIPATING",
"232": "SUCCESS",
"233": "NON_PARTICIPATING",
"234": "NON_PARTICIPATING",
"235": "NON_PARTICIPATING",
"236": "NON_PARTICIPATING",
"237": "NON_PARTICIPATING",
"238": "SUCCESS",
"239": "NON_PARTICIPATING",
"24": "SUCCESS",
"240": "NON_PARTICIPATING",
"241": "NON_PARTICIPATING",
"242": "SUCCESS",
"243": "NON_PARTICIPATING",
"244": "NON_PARTICIPATING",
"245": "NON_PARTICIPATING",
"246": "NON_PARTICIPATING",
"247": "SUCCESS",
"248": "NON_PARTICIPATING",
"249": "SUCCESS",
"25": "NON_PARTICIPATING",
"250": "SUCCESS",
"251": "NON_PARTICIPATING",
"252": "NON_PARTICIPATING",
"253": "NON_PARTICIPATING",
"254": "NON_PARTICIPATING",
"255": "NON_PARTICIPATING",
"26": "SUCCESS",
"27": "SUCCESS",
"28": "SUCCESS",
"29": "SUCCESS",
"3": "SUCCESS",
"30": "NON_PARTICIPATING",
"31": "SUCCESS",
"32": "NON_PARTICIPATING",
"33": "NON_PARTICIPATING",
"34": "SUCCESS",
"35": "SUCCESS",
"36": "SUCCESS",
"37": "SUCCESS",
"38": "SUCCESS",
"39": "NON_PARTICIPATING",
"4": "SUCCESS",
"40": "SUCCESS",
"41": "SUCCESS",
"42": "SUCCESS",
"43": "SUCCESS",
"44": "NON_PARTICIPATING",
"45": "SUCCESS",
"46": "SUCCESS",
"47": "SUCCESS",
"48": "NON_PARTICIPATING",
"49": "NON_PARTICIPATING",
"5": "SUCCESS",
"50": "NON_PARTICIPATING",
"51": "FAIL",
"52": "NON_PARTICIPATING",
"53": "SUCCESS",
"54": "FAIL",
"55": "SUCCESS",
"56": "SUCCESS",
"57": "NON_PARTICIPATING",
"58": "SUCCESS",
"59": "SUCCESS",
"6": "NON_PARTICIPATING",
"60": "SUCCESS",
"61": "NON_PARTICIPATING",
"62": "SUCCESS",
"63": "NON_PARTICIPATING",
"64": "SUCCESS",
"65": "SUCCESS",
"66": "SUCCESS",
"67": "SUCCESS",
"68": "SUCCESS",
"69": "SUCCESS",
"7": "NON_PARTICIPATING",
"70": "SUCCESS",
"71": "SUCCESS",
"72": "SUCCESS",
"73": "SUCCESS",
"74": "SUCCESS",
"75": "SUCCESS",
"76": "SUCCESS",
"77": "SUCCESS",
"78": "SUCCESS",
"79": "SUCCESS",
"8": "NON_PARTICIPATING",
"80": "SUCCESS",
"81": "SUCCESS",
"82": "FAIL",
"83": "SUCCESS",
"84": "SUCCESS",
"85": "SUCCESS",
"86": "SUCCESS",
"87": "SUCCESS",
"88": "SUCCESS",
"89": "SUCCESS",
"9": "SUCCESS",
"90": "SUCCESS",
"91": "NON_PARTICIPATING",
"92": "SUCCESS",
"93": "SUCCESS",
"94": "FAIL",
"95": "SUCCESS",
"96": "SUCCESS",
"97": "NON_PARTICIPATING",
"98": "SUCCESS",
"99": "SUCCESS"
},
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"block_list": [
5799307,
5799314,
5799336
],
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"inner_step": 64,
"intermediate_size": 5632,
"last_allreduce_block": 5798733,
"max_position_embeddings": 2048,
"mlp_bias": false,
"model_type": "llama",
"num_attention_heads": 32,
"num_hidden_layers": 22,
"num_key_value_heads": 4,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "float32",
"transformers_version": "4.39.3",
"use_cache": false,
"vocab_size": 32000
}