sharpenb's picture
7d4536029fe5e4db5cd98e13f8b4ecee4c431d6be297d1840f76e612907ee4c9
a463fa6 verified
raw
history blame
10.1 kB
{
"_name_or_path": "/tmp/models/tmpg2re1g5j/tmpxq2s2deg",
"activation_function": "gelu_new",
"all_reduce_scores": {
"0": "NON_PARTICIPATING",
"1": "NON_PARTICIPATING",
"10": "NON_PARTICIPATING",
"100": "NON_PARTICIPATING",
"101": "SUCCESS",
"102": "SUCCESS",
"103": "NON_PARTICIPATING",
"104": "NON_PARTICIPATING",
"105": "SUCCESS",
"106": "NON_PARTICIPATING",
"107": "SUCCESS",
"108": "NON_PARTICIPATING",
"109": "NON_PARTICIPATING",
"11": "NON_PARTICIPATING",
"110": "NON_PARTICIPATING",
"111": "NON_PARTICIPATING",
"112": "NON_PARTICIPATING",
"113": "SUCCESS",
"114": "NON_PARTICIPATING",
"115": "SUCCESS",
"116": "NON_PARTICIPATING",
"117": "NON_PARTICIPATING",
"118": "SUCCESS",
"119": "SUCCESS",
"12": "NON_PARTICIPATING",
"120": "SUCCESS",
"121": "SUCCESS",
"122": "SUCCESS",
"123": "SUCCESS",
"124": "NON_PARTICIPATING",
"125": "NON_PARTICIPATING",
"126": "NON_PARTICIPATING",
"127": "SUCCESS",
"128": "NON_PARTICIPATING",
"129": "SUCCESS",
"13": "NON_PARTICIPATING",
"130": "SUCCESS",
"131": "NON_PARTICIPATING",
"132": "SUCCESS",
"133": "NON_PARTICIPATING",
"134": "SUCCESS",
"135": "SUCCESS",
"136": "SUCCESS",
"137": "NON_PARTICIPATING",
"138": "NON_PARTICIPATING",
"139": "NON_PARTICIPATING",
"14": "NON_PARTICIPATING",
"140": "SUCCESS",
"141": "NON_PARTICIPATING",
"142": "SUCCESS",
"143": "NON_PARTICIPATING",
"144": "NON_PARTICIPATING",
"145": "NON_PARTICIPATING",
"146": "NON_PARTICIPATING",
"147": "NON_PARTICIPATING",
"148": "NON_PARTICIPATING",
"149": "SUCCESS",
"15": "NON_PARTICIPATING",
"150": "SUCCESS",
"151": "SUCCESS",
"152": "NON_PARTICIPATING",
"153": "SUCCESS",
"154": "NON_PARTICIPATING",
"155": "NON_PARTICIPATING",
"156": "SUCCESS",
"157": "NON_PARTICIPATING",
"158": "NON_PARTICIPATING",
"159": "SUCCESS",
"16": "NON_PARTICIPATING",
"160": "NON_PARTICIPATING",
"161": "NON_PARTICIPATING",
"162": "SUCCESS",
"163": "NON_PARTICIPATING",
"164": "SUCCESS",
"165": "NON_PARTICIPATING",
"166": "SUCCESS",
"167": "NON_PARTICIPATING",
"168": "NON_PARTICIPATING",
"169": "NON_PARTICIPATING",
"17": "SUCCESS",
"170": "SUCCESS",
"171": "NON_PARTICIPATING",
"172": "SUCCESS",
"173": "NON_PARTICIPATING",
"174": "SUCCESS",
"175": "NON_PARTICIPATING",
"176": "NON_PARTICIPATING",
"177": "NON_PARTICIPATING",
"178": "SUCCESS",
"179": "NON_PARTICIPATING",
"18": "NON_PARTICIPATING",
"180": "NON_PARTICIPATING",
"181": "NON_PARTICIPATING",
"182": "NON_PARTICIPATING",
"183": "NON_PARTICIPATING",
"184": "SUCCESS",
"185": "NON_PARTICIPATING",
"186": "NON_PARTICIPATING",
"187": "NON_PARTICIPATING",
"188": "NON_PARTICIPATING",
"189": "SUCCESS",
"19": "SUCCESS",
"190": "NON_PARTICIPATING",
"191": "NON_PARTICIPATING",
"192": "SUCCESS",
"193": "NON_PARTICIPATING",
"194": "SUCCESS",
"195": "SUCCESS",
"196": "NON_PARTICIPATING",
"197": "NON_PARTICIPATING",
"198": "NON_PARTICIPATING",
"199": "SUCCESS",
"2": "NON_PARTICIPATING",
"20": "NON_PARTICIPATING",
"200": "NON_PARTICIPATING",
"201": "NON_PARTICIPATING",
"202": "SUCCESS",
"203": "SUCCESS",
"204": "NON_PARTICIPATING",
"205": "SUCCESS",
"206": "SUCCESS",
"207": "NON_PARTICIPATING",
"208": "NON_PARTICIPATING",
"209": "NON_PARTICIPATING",
"21": "SUCCESS",
"210": "NON_PARTICIPATING",
"211": "SUCCESS",
"212": "NON_PARTICIPATING",
"213": "NON_PARTICIPATING",
"214": "NON_PARTICIPATING",
"215": "SUCCESS",
"216": "SUCCESS",
"217": "SUCCESS",
"218": "NON_PARTICIPATING",
"219": "NON_PARTICIPATING",
"22": "NON_PARTICIPATING",
"220": "SUCCESS",
"221": "NON_PARTICIPATING",
"222": "NON_PARTICIPATING",
"223": "NON_PARTICIPATING",
"224": "NON_PARTICIPATING",
"225": "NON_PARTICIPATING",
"226": "NON_PARTICIPATING",
"227": "SUCCESS",
"228": "SUCCESS",
"229": "NON_PARTICIPATING",
"23": "NON_PARTICIPATING",
"230": "NON_PARTICIPATING",
"231": "SUCCESS",
"232": "SUCCESS",
"233": "NON_PARTICIPATING",
"234": "SUCCESS",
"235": "SUCCESS",
"236": "NON_PARTICIPATING",
"237": "SUCCESS",
"238": "NON_PARTICIPATING",
"239": "SUCCESS",
"24": "NON_PARTICIPATING",
"240": "NON_PARTICIPATING",
"241": "NON_PARTICIPATING",
"242": "NON_PARTICIPATING",
"243": "NON_PARTICIPATING",
"244": "NON_PARTICIPATING",
"245": "NON_PARTICIPATING",
"246": "NON_PARTICIPATING",
"247": "NON_PARTICIPATING",
"248": "NON_PARTICIPATING",
"249": "NON_PARTICIPATING",
"25": "NON_PARTICIPATING",
"250": "SUCCESS",
"251": "NON_PARTICIPATING",
"252": "SUCCESS",
"253": "NON_PARTICIPATING",
"254": "NON_PARTICIPATING",
"255": "SUCCESS",
"26": "SUCCESS",
"27": "SUCCESS",
"28": "NON_PARTICIPATING",
"29": "NON_PARTICIPATING",
"3": "NON_PARTICIPATING",
"30": "NON_PARTICIPATING",
"31": "NON_PARTICIPATING",
"32": "SUCCESS",
"33": "NON_PARTICIPATING",
"34": "NON_PARTICIPATING",
"35": "NON_PARTICIPATING",
"36": "SUCCESS",
"37": "SUCCESS",
"38": "NON_PARTICIPATING",
"39": "NON_PARTICIPATING",
"4": "NON_PARTICIPATING",
"40": "NON_PARTICIPATING",
"41": "SUCCESS",
"42": "NON_PARTICIPATING",
"43": "NON_PARTICIPATING",
"44": "SUCCESS",
"45": "NON_PARTICIPATING",
"46": "SUCCESS",
"47": "SUCCESS",
"48": "NON_PARTICIPATING",
"49": "NON_PARTICIPATING",
"5": "NON_PARTICIPATING",
"50": "SUCCESS",
"51": "NON_PARTICIPATING",
"52": "SUCCESS",
"53": "NON_PARTICIPATING",
"54": "NON_PARTICIPATING",
"55": "NON_PARTICIPATING",
"56": "NON_PARTICIPATING",
"57": "NON_PARTICIPATING",
"58": "NON_PARTICIPATING",
"59": "SUCCESS",
"6": "SUCCESS",
"60": "SUCCESS",
"61": "SUCCESS",
"62": "NON_PARTICIPATING",
"63": "SUCCESS",
"64": "NON_PARTICIPATING",
"65": "NON_PARTICIPATING",
"66": "SUCCESS",
"67": "NON_PARTICIPATING",
"68": "NON_PARTICIPATING",
"69": "SUCCESS",
"7": "SUCCESS",
"70": "NON_PARTICIPATING",
"71": "NON_PARTICIPATING",
"72": "SUCCESS",
"73": "SUCCESS",
"74": "SUCCESS",
"75": "SUCCESS",
"76": "NON_PARTICIPATING",
"77": "NON_PARTICIPATING",
"78": "SUCCESS",
"79": "SUCCESS",
"8": "SUCCESS",
"80": "SUCCESS",
"81": "NON_PARTICIPATING",
"82": "NON_PARTICIPATING",
"83": "NON_PARTICIPATING",
"84": "SUCCESS",
"85": "NON_PARTICIPATING",
"86": "NON_PARTICIPATING",
"87": "SUCCESS",
"88": "SUCCESS",
"89": "SUCCESS",
"9": "NON_PARTICIPATING",
"90": "NON_PARTICIPATING",
"91": "NON_PARTICIPATING",
"92": "NON_PARTICIPATING",
"93": "NON_PARTICIPATING",
"94": "NON_PARTICIPATING",
"95": "NON_PARTICIPATING",
"96": "NON_PARTICIPATING",
"97": "SUCCESS",
"98": "SUCCESS",
"99": "SUCCESS"
},
"architectures": [
"GPTOptim"
],
"attn_pdrop": 0.1,
"auto_map": {
"AutoConfig": "configuration_gpt_optimized.GPTOptimConfig",
"AutoModelForCausalLM": "distributed/optimized-gpt2-2b--modeling_gpt_optimized.GPTOptim"
},
"block_size": 1024,
"bos_token_id": 50256,
"embd_pdrop": 0.1,
"eos_token_id": 50256,
"initializer_range": 0.02,
"layer_norm_epsilon": 1e-05,
"model_type": "gpt_optimized",
"n_embd": 1904,
"n_head": 34,
"n_inner": null,
"n_layer": 44,
"n_positions": 1024,
"quantization_config": {
"transformer.h.attn.c_attn": {
"offload_meta": false,
"scale_quant_params": null,
"weight_quant_params": {
"axis": 0,
"channel_wise": true,
"group_size": 64,
"nbits": 8,
"optimize": true,
"round_zero": false,
"view_as_float": false
},
"zero_quant_params": {
"channel_wise": false,
"group_size": null,
"nbits": 8,
"optimize": false
}
},
"transformer.h.attn.c_proj": {
"offload_meta": false,
"scale_quant_params": null,
"weight_quant_params": {
"axis": 0,
"channel_wise": true,
"group_size": 64,
"nbits": 8,
"optimize": true,
"round_zero": false,
"view_as_float": false
},
"zero_quant_params": {
"channel_wise": false,
"group_size": null,
"nbits": 8,
"optimize": false
}
},
"transformer.h.mlp.c_fc": {
"offload_meta": false,
"scale_quant_params": null,
"weight_quant_params": {
"axis": 0,
"channel_wise": true,
"group_size": 64,
"nbits": 8,
"optimize": true,
"round_zero": false,
"view_as_float": false
},
"zero_quant_params": {
"channel_wise": false,
"group_size": null,
"nbits": 8,
"optimize": false
}
},
"transformer.h.mlp.c_proj": {
"offload_meta": false,
"scale_quant_params": null,
"weight_quant_params": {
"axis": 0,
"channel_wise": true,
"group_size": 64,
"nbits": 8,
"optimize": true,
"round_zero": false,
"view_as_float": false
},
"zero_quant_params": {
"channel_wise": false,
"group_size": null,
"nbits": 8,
"optimize": false
}
}
},
"reorder_and_upcast_attn": false,
"resid_pdrop": 0.1,
"scale_attn_by_inverse_layer_idx": false,
"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,
"torch_dtype": "float32",
"transformers_version": "4.44.0",
"use_cache": true,
"vocab_size": 50257
}